System

Abstract

A system is provided.SOLUTION: A system, comprising: means for collecting voice data input by a user; means for transmitting the voice data to a server; means for converting the voice data into text data by the server; means for translating the text data into a specified language by the server; means for converting the translated text data into voice data by the server; and means for outputting the translated voice data to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] The present invention aims to eliminate language barriers when communicating between multiple languages ​​in real time. Conventional translation applications have problems such as complicated operation, low translation accuracy, and a large time lag, making smooth dialogue difficult. Furthermore, when an interpreter is required, there are issues with cost and reliability. The objective of this invention is to provide a system that solves these problems and enables seamless and accurate communication between users and the other party. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means: a system including means for collecting voice data input by a user, means for transmitting the voice data to a server, means for the server to convert the voice data into character data, means for translating the character data into a specified language, means for converting the translated character data into voice data, and means for outputting the translated voice data to the user. This allows the entire process, from voice input to translation and voice output, to be carried out in real time, enabling smooth communication between the user and the other party.

[0006] A "user" is an entity that uses the system to input voice data and output translated voice data.

[0007] "Voice data" refers to the digital signal of the voice input by the user, and is the subject of voice recognition and translation processing.

[0008] A "server" is a computer system or network that receives, processes, converts, and translates audio data.

[0009] "Character data" is text information in a language converted by speech recognition.

[0010] "Translation" refers to the process of converting character data expressed in one language into another language.

[0011] A "voice recognition engine" is software or hardware that analyzes voice data and converts it into text-format character data.

[0012] A "natural language processing engine" is software or hardware that analyzes character data and translates it into a specified language.

[0013] A "speech synthesis engine" is software or hardware that analyzes text data and outputs it as voice data. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram illustrating a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] The multilingual machine translation system of the present invention is implemented as a smartphone application. A specific embodiment of the system will be described below.

[0036] System Overview:

[0037] The system combines speech recognition, natural language processing, and speech synthesis technologies to support real-time communication between users and speakers of other languages. Users use their smartphones to input what they want to say, and the system instantly translates and outputs the speech.

[0038] Specific steps for implementation:

[0039] User Action:

[0040] After launching the app, the user sets the "language they speak," the "language they want to convert to," and the "desired voice." The user then speaks into the smartphone's microphone, for example, saying, "Hello, what's your name?" The smartphone device collects this voice data, converts it into digital format, and sends it to a server.

[0041] Server Action:

[0042] The server passes the received voice data to a voice recognition engine, which first converts the voice into text data. This text data is then passed to a natural language processing engine, which translates it into the specified language. The translated text data is then sent to a speech synthesis engine, which generates voice data based on the "desired voice." The generated voice data is then sent from the server to the device.

[0043] The resulting output:

[0044] The terminal plays back the received voice data and lets the user and the other party hear it. For example, a voice in English saying "Hello, what is your name?" is played back.

[0045] Response from the other party:

[0046] If the other person says "My name is John" in English, the smartphone device collects this speech and sends it back to the server. The server then performs speech recognition, natural language processing, and speech synthesis in the same way, and outputs the translated speech data to the user. This allows the user to hear the Japanese audio of "My name is John" on their smartphone.

[0047] Examples:

[0048] 1. User and partner initial settings

[0049] User: Launches the app and selects "Japanese" as the spoken language and "English" as the target language. Selects "Male voice" as the desired voice.

[0050] Other person: Speak in English.

[0051] 2. User utterances

[0052] User: Say "Hello, what's your name?"

[0053] Terminal: Sends audio data to the server.

[0054] Server: The speech recognition engine converts the speech into text and saves it in the log as "Hello, what is your name?". The natural language processing engine translates it into "Hello, what is your name?". The speech synthesis engine converts the translated text into speech. The speech data is sent to the device.

[0055] Device: Play the received audio, "Hello, what is your name?"

[0056] 3. The other person's response

[0057] Other person: Say, "My name is John."

[0058] Terminal: Sends the other party's voice data to the server.

[0059] Server: The speech recognition engine converts the speech into text and saves it in the log as "My name is John." The natural language processing engine translates it as "My name is John." The speech synthesis engine converts the converted text into speech. The speech data is sent to the device.

[0060] Device: Play the received audio, "My name is John."

[0061] This allows the user and the other party to communicate seamlessly in real time. The present invention achieves high translation accuracy with almost no time lag and simple operation.

[0062] The processing flow will be explained below.

[0063] Step 1:

[0064] User: Launch the app and set the "language you speak," "language to convert to," and "desired voice" on the settings screen. For example, set "Japanese" as the language you speak, "English" as the language you want to convert to, and "male voice" as the desired voice.

[0065] Step 2:

[0066] Terminal: Saves the user's settings and enters a standby state for voice input.

[0067] Step 3:

[0068] User: Begins speaking into the smartphone microphone. For example, "Hello, what's your name?"

[0069] Step 4:

[0070] Terminal: Collects the user's voice as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[0071] Step 5:

[0072] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "Hello, what's your name?". The converted text data is recorded in a log.

[0073] Step 6:

[0074] Server: Passes the text data to the natural language processing engine. The natural language processing engine translates "Hello, what is your name?" into "Hello, what is your name?" and records the translation result in a log.

[0075] Step 7:

[0076] Server: Passes the translated text data to the speech synthesis engine. The speech synthesis engine converts "Hello, what is your name?" into audio data in a "male voice." The generated audio data is prepared for transmission to the device.

[0077] Step 8:

[0078] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "Hello, what is your name?"

[0079] Step 9:

[0080] Other person: Respond in English. For example, say, "My name is John."

[0081] Step 10:

[0082] Terminal: Collects the other party's voice data as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[0083] Step 11:

[0084] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "My name is John." The converted text data is recorded in a log.

[0085] Step 12:

[0086] Server: Passes the text data to the natural language processing engine. The natural language processing engine translates "My name is John." into "My name is John." The translation result is recorded in a log.

[0087] Step 13:

[0088] Server: Passes the translated text data to the speech synthesis engine. The speech synthesis engine converts "My name is John" into audio data in a "male voice." The generated audio data is prepared for transmission to the device.

[0089] Step 14:

[0090] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "My name is John."

[0091] Example 1

[0092] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0093] Existing multilingual translation systems often experience a time lag between speech input and speech output of the translation results, which can hinder smooth real-time communication between the user and the other party. Furthermore, the low speech quality and translation accuracy make accurate communication difficult. Furthermore, the inability to generate speech in the user's desired voice makes it difficult to provide a natural conversational experience.

[0094] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0095] In this invention, the server includes means for converting voice data input by a user into character data, means for translating the character data into a specified language, and means for converting the translated character data into voice data, thereby enabling voice output in a voice desired by the user and realizing highly accurate multilingual translation in real time.

[0096] "User" refers to the person who inputs voice data and receives the translation results.

[0097] "Voice data" refers to data that has been converted into digital form from the user's spoken voice.

[0098] "Server" refers to a central control device that processes voice data, converts it to text data, translates it, and converts it back to voice data.

[0099] "Character data" refers to data that has been converted from audio data into text format.

[0100] "Translation" refers to the process of converting character data into character data in a specified different language.

[0101] The "specified language" refers to the language into which the user wants to translate the voice data.

[0102] "Voice output" refers to the process of converting translated text data into voice data and providing it to the user.

[0103] "Digital data" refers to the digital format data used when transmitting audio data to a server.

[0104] "Desired voice" refers to the voice characteristics (e.g., gender and tone of voice) that a user uses when outputting voice.

[0105] "Terminal" refers to a device (e.g., a smartphone) that allows a user to input voice data and receive and play back the voice data of the translation result.

[0106] "Log" refers to data storage that records the voice data processed by the server and the translation results.

[0107] The multilingual machine translation system of the present invention is implemented as a smartphone app. This system translates speech data spoken by a user into other languages ​​in real time and outputs the speech in the user's desired voice, thereby smoothly supporting communication between the user and speakers of other languages. Specific embodiments of this system are described in detail below.

[0108] Overall system picture

[0109] This system combines speech recognition, natural language processing, and speech synthesis technologies, allowing users to input what they want to say by voice using their smartphone, and the system instantly translates and outputs the speech.

[0110] Hardware and software used

[0111] Hardware

[0112] Smartphone (terminal): Used for user voice input and voice output

[0113] Server: A central control unit used to process voice data.

[0114] software

[0115] Speech recognition engine: Google Cloud Speech-to-Text

[0116] Natural language processing engine: Google Cloud Translation

[0117] Speech synthesis engine: Google Cloud Text-to-Speech

[0118] Specific processing of the system

[0119] 1. User Initial Settings

[0120] The user starts the smartphone app and sets the "language they speak," the "language they want to convert to," and the "desired voice." These settings can be made on the app's settings screen.

[0121] 2. Audio input and digital conversion

[0122] Based on the user's settings, when the user speaks into the smartphone's microphone, the device converts the voice data into digital format and sends it to the server.

[0123] 3. Processing of audio data by the server

[0124] The server processes the received audio data as follows:

[0125] Speech recognition: Use the Google Cloud Speech-to-Text API to convert voice data into text data.

[0126] Natural Language Processing: Translates text data into a specified language using the Google Cloud Translation API.

[0127] Speech synthesis: Using the Google Cloud Text-to-Speech API, the translated text data is converted into voice data based on the "voice of desire."

[0128] 4. Data transmission and audio output

[0129] The server sends the generated voice data to the smartphone, which then plays it back, allowing the user to hear the translated voice.

[0130] Examples of specific examples and prompts

[0131] Specific examples

[0132] User: Launches the smartphone app, sets "Japanese" as the spoken language, "English" as the target language, and selects "male voice" as the desired voice.

[0133] User: Say "Hello, what's your name?"

[0134] Terminal: Sends audio data to the server.

[0135] Server: (Speech recognition) Convert "Hello, what's your name?" into text data.

[0136] Server: (Natural Language Processing) Translates to "Hello, what is your name?"

[0137] Server: (Speech synthesis) Generates English speech in a male voice.

[0138] Server: Sends audio data to the terminal.

[0139] Device: Play "Hello, what is your name?"

[0140] Other person: "My name is John."

[0141] Device: Sends audio to the server.

[0142] Server: (Speech recognition) Convert "My name is John." into text data.

[0143] Server: (Natural Language Processing) Translates to "My name is John."

[0144] Server: (Speech synthesis) Generates Japanese speech.

[0145] Server: Sends audio data to the terminal.

[0146] Device: Play "My name is John."

[0147] Prompt Sentence Examples

[0148] "Please translate the following phrase and play it aloud: Hello, what's your name?"

[0149] Please translate Japanese into English.

[0150] Through the above-described processing, the present invention realizes real-time, highly accurate multilingual translation and supports smooth communication between users and speakers of other languages.

[0151] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0152] Step 1:

[0153] The user launches the smartphone app and sets the "language they speak," the "language they want to translate into," and the "desired voice." This allows the user to specify which language the system will speak and which language it will translate into. For example, they can set the translation from Japanese to English and output it in a male voice.

[0154] Step 2:

[0155] The user speaks into the device's microphone, generating specific input voice data such as "Hello, what's your name?" The device converts this voice into digital data. Here, the input is an analog voice signal, and the output is digital voice data.

[0156] Step 3:

[0157] The terminal sends digital audio data to the server. This communication is carried out over a network. The input from the terminal is digital audio data, and the output to the server is also digital audio data.

[0158] Step 4:

[0159] The server converts the received voice data into text data using the Google Cloud Speech-to-Text API. The input is digital voice data, which is analyzed and the output is the text data "Hello, what's your name?"

[0160] Step 5:

[0161] The server translates the text data into the specified language using the Google Cloud Translation API. The input is Japanese text data, and the output is English text data "Hello, what is your name?". At this time, the server recognizes the language of the text and converts it into the specified language.

[0162] Step 6:

[0163] The server uses the Google Cloud Text-to-Speech API to convert the translated text data into audio data based on the desired voice. The input is the English text data "Hello, what is your name?", and the output is audio data generated in a male voice.

[0164] Step 7:

[0165] The server sends the generated voice data to the terminal. The input is the generated voice data, and the output is data transfer to the terminal. This is also done over the network.

[0166] Step 8:

[0167] The terminal plays the received voice data and lets the user and the other party listen. The terminal's input is the received voice data, and the output is the voice "Hello, what is your name?" played through the speaker.

[0168] Step 9:

[0169] The other person speaks in English into the device's microphone, "My name is John." This voice data is generated. The other person's input is an analog voice signal, and the device's output is digital voice data.

[0170] Step 10:

[0171] The terminal transmits the other party's digital voice data to the server. The input is the other party's digital voice data, and the output is the data transmitted to the server.

[0172] Step 11:

[0173] The server converts the received voice data back into text data using the Google Cloud Speech-to-Text API. The input is the other person's digital voice data, and the output is the text data "My name is John."

[0174] Step 12:

[0175] The server uses the Google Cloud Translation API to translate the text data into the specified language. The input is English text data, and the output is Japanese text data: "My name is John."

[0176] Step 13:

[0177] The server uses the Google Cloud Text-to-Speech API to convert the translated text data into audio data. The input is the Japanese text data "My name is John." and the output is audio data.

[0178] Step 14:

[0179] The server sends the generated voice data to the terminal. The input is the generated voice data, and the output is the data transfer to the terminal.

[0180] Step 15:

[0181] The terminal plays the received voice data and lets the user hear it. The input is the received voice data, and the output is the voice "My name is John" played through the speaker.

[0182] This allows for seamless real-time communication between users and speakers of other languages.

[0183] (Application example 1)

[0184] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0185] In conventional food delivery services, when the delivery person and the customer speak different languages, communication problems arise, making it difficult to deliver smoothly. This makes it difficult to immediately respond to the exact delivery location or special requests of the customer, and there is a need to improve the quality of the service.

[0186] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0187] In this invention, the server includes means for collecting voice data input by a user, means for transmitting the voice data to the server, means for the server to convert the voice data into character data, means for the server to translate the character data into a specified language, means for the server to convert the translated character data into voice data, means for outputting the translated voice data to the user, means for synthesizing the translation of the voice data into a specified language, and means for supporting communication between food delivery personnel and customers, thereby enabling smooth real-time communication between delivery personnel and customers who speak different languages.

[0188] "User-input voice data" refers to data that is collected by a device as an electrical signal when a user speaks through an input device such as a microphone.

[0189] "Means for transmitting to a server" refers to the general technology and protocols for transferring collected voice data to a server via a communications network.

[0190] "Means for converting into text data" refers to a process or engine that uses speech recognition technology to convert collected voice data into corresponding text data.

[0191] "Means for translating into a specified language" refers to technology for converting text data into a different language using a natural language processing engine.

[0192] "Means for converting into voice data" refers to a process for converting the translated text data into voice data using voice synthesis technology that synthesizes natural human speech from text.

[0193] "Means for outputting translated audio data to the user" refers to technology for playing the audio data through an output device such as a speaker.

[0194] "Means for supporting communication between food delivery personnel and customers" refers to all technologies that utilize a multilingual automatic translation system to enable real-time communication between delivery personnel and customers who speak different languages.

[0195] This invention is a multilingual automatic translation system that supports real-time communication between a user and people who speak different languages, and has an embodiment specialized for food delivery services. Specific processing and each means of this embodiment will be described below.

[0196] System Overview:

[0197] The system provides real-time multilingual translation between delivery personnel and customers, facilitating smooth communication. The system is realized by combining speech recognition, natural language processing, and speech synthesis technologies. The main hardware and software used include smartphones, servers, speech recognition engines (e.g., Google Speech-to-Text API), natural language processing engines (e.g., Google Translate API), and speech synthesis engines (e.g., gTTS).

[0198] Program processing:

[0199] Voice collection and recognition:

[0200] The device (smartphone) collects Japanese voice data input by the user. The device uses a built-in microphone to record the user's voice and sends it to the server as digital voice data. The server then uses the Google Speech-to-Text API to convert this voice data into text data.

[0201] Text translation:

[0202] The server translates the converted text data into a specified language, for example, English, using the Google Translate API. The translated text data is temporarily stored on the server.

[0203] Text-to-Speech:

[0204] The translated text data is converted into audio data using gTTS (Google Text-to-Speech), which is then sent back to the device and played through the speaker.

[0205] Examples:

[0206] User voice input:

[0207] For example, a delivery person might say in Japanese, "May I confirm your address?" The device collects this voice and sends it to the server.

[0208] Server process:

[0209] The speech recognition engine converts the speech into text and saves it in the log as "May I confirm your address?"

[0210] A natural language processing engine translates this text into English as "Can I confirm the address?"

[0211] A speech synthesis engine converts the translated text into speech.

[0212] Audio Output:

[0213] The device plays a translated English audio message that the delivery person can instantly understand in the customer's language, facilitating the confirmation of the appropriate delivery location.

[0214] Examples of prompts:

[0215] A user launches the app, sets "Japanese" as the primary language and "English" as the target language, and then says "May I confirm your address?" in Japanese. The app recognizes this, translates it into English, and plays it back aloud.

[0216] In this way, the barrier of multilingual communication in food delivery services is resolved, enabling smooth communication between delivery personnel and customers.

[0217] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0218] Step 1:

[0219] The user launches the app and provides voice input. The user launches the app on their smartphone and sets "Japanese" as the main language and "English" as the translation language. They then speak in Japanese, for example, "May I confirm your address?" The device's microphone collects this voice and processes it as digital voice data. The input in this step is the user's voice, and the output is digital voice data.

[0220] Step 2:

[0221] The device sends the collected audio data to the server. The device uploads this digital audio data to the server via an internet connection. The input in this step is the digital audio data, and the output is the audio data sent to the server.

[0222] Step 3:

[0223] The server uses a speech recognition engine to convert the transmitted voice data into text data. Specifically, the server calls the Google Speech-to-Text API to convert the voice data into text format. The input in this step is voice data, and the output is text data such as "May I confirm your address?"

[0224] Step 4:

[0225] The server passes the converted text data to a natural language processing engine and translates it into the specified language. Specifically, the server uses the Google Translate API to convert the text data "May I confirm the address?" into the English text "Can I confirm the address?". The input in this step is the text data, and the output is the translated text data.

[0226] Step 5:

[0227] The server uses a speech synthesis engine to convert the translated text data into audio data. Specifically, the server uses gTTS (Google Text-to-Speech) to convert the translated English text into an audio file. The input in this step is the translated text data, and the output is audio data.

[0228] Step 6:

[0229] The server sends the generated voice data to the terminal. The server uploads the generated voice data to the terminal via an internet connection. The input in this step is the voice data, and the output is the voice data sent to the terminal.

[0230] Step 7:

[0231] The device plays the received audio data. Specifically, it uses the device's speaker to play the translated English audio. By asking the user, "Can I confirm the address?", the delivery person can immediately understand it in the customer's language. The input in this step is audio data, and the output is audio that the user can hear.

[0232] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0233] The multilingual machine translation system of the present invention combines speech recognition, natural language processing, speech synthesis, and emotion recognition technologies to support real-time communication between users and speakers of other languages. Users input what they want to say using their smartphone, and the system instantly translates and outputs voice based on the emotion.

[0234] System Overview:

[0235] This system collects the user's voice data, performs speech recognition and emotion recognition on the server, and translates it using natural language processing. Furthermore, it synthesizes speech based on the emotion recognition results and outputs speech that reflects the appropriate emotion to the user. This not only overcomes language barriers, but also makes it possible to accurately convey emotions.

[0236] Specific steps for implementation:

[0237] User Action:

[0238] The user launches the app and sets the "language they speak," "language to convert to," "desired voice," and "emotion recognition settings" on the settings screen. The user then speaks into the smartphone's microphone, saying, for example, "Hello, what's your name?"

[0239] Audio data collection:

[0240] The terminal collects the user's voice data as a digital signal, converts it into voice data, and transmits it to a server via the Internet.

[0241] Speech and emotion recognition:

[0242] The server passes the received voice data to a voice recognition engine and an emotion engine. The voice recognition engine converts the voice data into text data, and the emotion engine recognizes the user's emotion (e.g., joy, sadness, anger, etc.).

[0243] Translation process:

[0244] The recognized text data is passed to a natural language processing engine and translated into the specified language. For example, "Hello, what is your name?" is translated into "Hello, what is your name?"

[0245] Emotion-based speech synthesis:

[0246] The server sends the translated text data and the recognized emotion data to a speech synthesis engine. The speech synthesis engine adjusts the translation result based on the emotion and generates speech data that reflects the appropriate emotion. For example, if the user is happy, the tone of the voice will also be adjusted to express happiness.

[0247] The resulting output:

[0248] The device receives the voice data from the server and plays it back through the smartphone speaker. The voice that is played back is "Hello, what is your name?", and reflects the user's emotions.

[0249] Response from the other party:

[0250] The other person responds in English, for example, "My name is John."

[0251] Processing the other party's voice data:

[0252] The device collects the other party's voice data and sends it back to the server. The server then performs speech recognition, natural language processing, and speech synthesis in the same way, and outputs the translated and emotion-reflected voice data to the user. For example, a voice saying "My name is John" is played back with the appropriate emotion.

[0253] Examples:

[0254] 1. User and partner initial settings

[0255] User: Launch the app and set "Japanese" as the speaking language, "English" as the target language, and "Male Voice" as the desired voice. Set emotion recognition to "On."

[0256] Other person: Speak in English.

[0257] 2. User utterances

[0258] User: Say "Hello, what's your name?"

[0259] Terminal: Sends audio data to the server.

[0260] Server: The speech recognition engine converts the speech into text data. It saves the message in the log as "Hello, what is your name?" The emotion engine recognizes "joy." The natural language processing engine translates it into "Hello, what is your name?" The speech synthesis engine converts it into English speech that expresses "joy." The speech data is sent to the device.

[0261] Device: Play the received audio. "Hello, what is your name?"

[0262] 3. The other person's response

[0263] Other person: Say, "My name is John."

[0264] Terminal: Sends the other party's voice data to the server.

[0265] Server: The speech recognition engine converts the speech into text data. It saves the text in the log as "My name is John." The emotion engine recognizes "neutral." The natural language processing engine translates it to "My name is John." The speech synthesis engine converts it into Japanese speech that includes "neutral." The speech data is sent to the device.

[0266] Device: Play the received audio: "My name is John."

[0267] This allows users and their partners to communicate seamlessly in real time, reflecting their emotions. This invention is a groundbreaking system in that it goes beyond simple language translation and also enables the transmission of emotions.

[0268] The processing flow will be explained below.

[0269] Step 1:

[0270] User: Launch the app and set the "language you speak," "language you want to convert to," "desired voice," and "emotion recognition settings" on the settings screen. For example, set "Japanese" as the language you speak, "English" as the language you want to convert to, "male voice" as the desired voice, and turn on "emotion recognition."

[0271] Step 2:

[0272] Terminal: Saves the user's settings and enters a standby state for voice input.

[0273] Step 3:

[0274] User: Begins speaking into the smartphone microphone. For example, "Hello, what's your name?"

[0275] Step 4:

[0276] Terminal: Collects the user's voice as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[0277] Step 5:

[0278] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "Hello, what's your name?". The converted text data is recorded in a log.

[0279] Step 6:

[0280] Server: Passes the text data to the natural language processing engine and translates it into the specified language. The natural language processing engine translates "Hello, what is your name?" into "Hello, what is your name?". The translation result is recorded in a log.

[0281] Step 7:

[0282] Server: Passes the voice data to the emotion engine and analyzes the user's emotion. The emotion engine generates an emotion tag from the voice, recognizing "joy" in this example, and records the result in a log.

[0283] Step 8:

[0284] Server: Passes the translated text data and emotion tag to the speech synthesis engine. The speech synthesis engine converts "Hello, what is your name?" into voice data in a "male voice" that reflects the emotion of "joy." The generated voice data is then prepared for transmission to the device.

[0285] Step 9:

[0286] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "Hello, what is your name?", with a tone that expresses "joy."

[0287] Step 10:

[0288] Other person: Respond in English. For example, say, "My name is John."

[0289] Step 11:

[0290] Terminal: Collects the other party's voice data again as a digital signal and sends it to a server via the Internet.

[0291] Step 12:

[0292] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "My name is John." The converted text data is recorded in a log.

[0293] Step 13:

[0294] Server: Passes the text data to the natural language processing engine and translates it into the specified language. The natural language processing engine translates "My name is John." to "My name is John." The translation result is recorded in a log.

[0295] Step 14:

[0296] Server: Passes the voice data to the emotion engine and analyzes the other person's emotion. The emotion engine generates an emotion tag from the voice, recognizing "neutral" in this example. Records the result in a log.

[0297] Step 15:

[0298] Server: Passes the translated text data and emotion tag to the speech synthesis engine. The speech synthesis engine converts "My name is John" into voice data in a "male voice" that reflects a "neutral" emotion. The generated voice data is then prepared for transmission to the device.

[0299] Step 16:

[0300] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "My name is John," in a tone that expresses "neutrality."

[0301] This step enables users and other parties to overcome language barriers and achieve real-time communication that accurately conveys emotions.

[0302] Example 2

[0303] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0304] Conventional multilingual translation systems simply translate languages ​​and have difficulty conveying emotional nuances. As a result, they are unable to accurately convey emotions, which play an important role in communication, and are therefore less convenient for actual dialogue. Furthermore, they have the problem of being difficult to translate and reflect emotions in real time, which hinders smooth communication.

[0305] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0306] In this invention, the server includes a means for converting voice data into text data, a means for recognizing emotions in the text data, and a means for converting the translated text data into voice data that reflects the emotions, thereby enabling accurate conveyance of emotions and smooth real-time multilingual communication.

[0307] "Audio data" refers to sound information input by a user through the microphone of a terminal, converted into a digital signal.

[0308] A "server" is a central processing unit that receives voice data and performs various processes such as preprocessing, voice recognition, emotion recognition, translation, and voice synthesis.

[0309] "Character data" is data in text format that has been converted from voice data by a voice recognition engine.

[0310] "Translation" is the process of converting character data into a specified language.

[0311] "Emotion recognition" is a technology that analyzes and identifies emotions (joy, surprise, sadness, etc.) from a user's voice data.

[0312] "Speech synthesis" is a technology that artificially generates natural-sounding speech based on character data and emotion recognition results.

[0313] A "terminal" is a digital device used by a user, such as a smartphone or tablet.

[0314] The multilingual machine translation system of the present invention combines speech recognition, natural language processing, speech synthesis, and emotion recognition technologies to support real-time communication between users and speakers of other languages. This system collects speech data provided by users, translates it through processing on the server, and generates and outputs speech that reflects appropriate emotions.

[0315] Hardware and Software Configuration

[0316] The main components of the system are:

[0317] 1. Device (smartphone or tablet):

[0318] A microphone that collects audio data

[0319] A function that converts audio data into a digital signal

[0320] Ability to send data to a server via the Internet

[0321] A speaker that plays back audio data received from the server

[0322] 2. Server:

[0323] Speech recognition engine (e.g., Google Cloud Speech-to-Text): converts voice data into text data

[0324] Emotion recognition engine (e.g., IBM Watson Tone Analyzer): Analyzes emotions from voice data

[0325] Natural language processing engine (e.g., Google Cloud Translation API): Translates text data into a specified language

[0326] Speech synthesis engine (e.g., Amazon Polly): Generates speech based on translated text data and emotion data.

[0327] Examples of the invention

[0328] Example 1:

[0329] The user launches the app and configures the following settings on the settings screen:

[0330] "Languages ​​I speak": "Japanese"

[0331] "Language to convert to": "English"

[0332] "Voice of Hope": "Male Voice"

[0333] "Emotion Recognition Settings": "On"

[0334] Next, the user speaks into the smartphone's microphone, saying, "Look at that dog!" The device collects this voice data as a digital signal and sends it to a server via the Internet. On the server, a voice recognition engine converts the voice data into "text data" and generates the text data "Look at that dog!" The emotion recognition engine then recognizes the user's emotion as "surprise" from the text data.

[0335] The server's natural language processing engine translates the text data into the specified language (English) and obtains the translation result "Look at that dog!". The speech synthesis engine generates English speech that reflects the appropriate emotion based on this translation result and the emotion of "surprise." The server then sends the final generated speech data to the device, and the surprised voice "Look at that dog!" is played from the device's speaker.

[0336] Example 2:

[0337] When the other person responds with "Yeah, it's really cute!", the device converts the other person's voice data into a digital signal and sends it to the server. The server uses a speech recognition engine to convert "Yeah, it's really cute!" into text data, and an emotion recognition engine recognizes the emotion "neutral." The natural language processing engine translates this to "Yeah, it's really cute!", and a speech synthesis engine generates Japanese speech that reflects a neutral tone. Finally, the generated voice data is sent to the device, and the speaker plays back the voice "Yeah, it's really cute!"

[0338] Example prompt sentence:

[0339] "The user launches the app and sets Japanese as the language they speak and English as the translation language. The user says, "Hello, what is your name?" The voice data is sent to the server, the speech recognition engine converts it to text, and the emotion engine recognizes "joy." The text data is translated by the NLP engine into "Hello, what is your name?" and the speech synthesis engine generates English speech that reflects joy. The audio is then played on the user's device."

[0340] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0341] Step 1:

[0342] The user launches the app and enters the language they speak, the language they want to translate into, the desired voice, and emotion recognition settings on the settings screen. Specifically, they set "Japanese" as the language they speak and "English" as the target language, then select a "male voice" and turn emotion recognition "on." Once the settings are complete, the user taps the "Save Settings" button.

[0343] Input: User settings

[0344] Output: The app with the settings saved

[0345] Step 2:

[0346] After completing the setup, the user speaks into the smartphone's microphone, for example, saying, "Look at that dog!" The device converts the voice data through the microphone into a digital signal and sends this data to a server via the Internet.

[0347] Input: User's voice

[0348] Output: Audio data sent to the server

[0349] Step 3:

[0350] The server receives the voice data and passes it to a speech recognition engine. The speech recognition engine (e.g., Google Cloud Speech-to-Text) converts the voice data into text data. For example, it generates text data such as "Look at that dog!" The server then passes the text data to an emotion recognition engine (e.g., IBM Watson Tone Analyzer), which recognizes the user's emotion as "surprise" from the text data.

[0351] Input: Audio data

[0352] Output: Text data and emotion data

[0353] Step 4:

[0354] The server passes the generated text data and emotion data to a natural language processing engine. The natural language processing engine (e.g., Google Cloud Translation API) translates this text data into the specified language (English in this case). For example, "Look at that dog!" is translated to "Look at that dog!"

[0355] Input: Text data and emotion data

[0356] Output: Translated text data (English)

[0357] Step 5:

[0358] The server sends the translated text data and emotion data to a speech synthesis engine. The speech synthesis engine (e.g., Amazon Polly) generates voice data that reflects the user's emotion based on the translation results. For example, voice data that expresses surprise and says, "Look at that dog!"

[0359] Input: Translated text data and emotion data

[0360] Output: Emotion-reflecting audio data

[0361] Step 6:

[0362] The server sends the generated voice data to the device, which then plays the received voice data through the smartphone speaker, allowing the user to hear the surprised voice saying, "Look at that dog!"

[0363] Input: Audio data

[0364] Output: Audio played through the speaker

[0365] Step 7:

[0366] The other person responds to the user's speech, for example, "Yeah, it's really cute!" The device collects the other person's voice data and sends it back to the server.

[0367] Input: Other party's voice

[0368] Output: Audio data sent to the server

[0369] Step 8:

[0370] The server then passes the received voice data back to the voice recognition engine, which converts it into text data. For example, the text data generated is "Yeah, it's really cute!" The server then passes the text data to the emotion recognition engine, which analyzes the emotion and determines that it is "neutral." The translation engine translates the text data into Japanese, obtaining the translation result "Yeah, it's really cute!" Finally, the speech synthesis engine generates Japanese speech based on this translation result and the "neutral" emotion. The device then plays back the received voice data, allowing the user to hear the voice saying "Yeah, it's really cute!"

[0371] Input: Voice data of the other party

[0372] Output: Emotionally-reflected Japanese speech data

[0373] (Application example 2)

[0374] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0375] In modern factories, it is becoming increasingly common for multinational workers to work together, which can lead to communication issues between workers who speak different languages. This can lead to reduced work efficiency and potential misunderstandings and work errors. Furthermore, in addition to simple language translation, there are also issues with emotions not being conveyed when giving work instructions or reporting, which can lead to a lack of understanding of the urgency and importance of instructions. A system that can solve these problems is needed.

[0376] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0377] In this invention, the server includes means for collecting voice data input by a user, means for transmitting the voice data to the server, means for converting the voice data into character data by the server, means for translating the character data into a specified language by the server, means for converting the translated character data into voice data by the server, means for recognizing emotions from the voice data and the character data by the server, means for adjusting the translated voice data based on emotions, and means for outputting the translated voice data to a user. This enables real-time communication between workers who speak different languages ​​that takes emotions into consideration.

[0378] "Voice data" refers to digital signals that collect end-user speech in audio form.

[0379] A "server" is a central computer system that processes, converts, and translates voice data over a network.

[0380] "Character data" is information in text format converted from voice data.

[0381] "Translation" is the process of converting text data from one language into another.

[0382] "Emotion recognition" is a technology that identifies a speaker's emotions (e.g., joy, sadness, anger, etc.) from audio data and text data.

[0383] "Speech synthesis" is a technology that generates speech based on text data.

[0384] "Real-time" means that processing occurs almost immediately, with minimal time delay.

[0385] "Modification" is the process of changing the tone and intonation of speech data based on perceived emotion.

[0386] "Log" means data storage for recording and saving processed data.

[0387] The present invention is a multilingual automatic translation system that translates what a user says into other languages ​​in real time and outputs the translated content while reflecting the user's emotions. Specific embodiments for carrying out the present invention will be described below.

[0388] Program Overview

[0389] The server is composed of multiple components, including a speech recognition engine, emotion recognition engine, translation engine, and speech synthesis engine. The user's device (for example, a smartphone or robot) collects voice data and sends it to the server. When the server receives the voice data, it performs speech recognition and converts it into text data. It then performs emotion recognition to identify the user's emotion. The translation engine translates the text data into the specified language, and the speech synthesis engine generates speech based on the translated text data and emotion data. Finally, the generated voice data is sent to the user's device and played back from the device.

[0390] Hardware and software used

[0391] 1. Hardware:

[0392] User terminal: devices such as smartphones and robots

[0393] Server: High-performance computing server

[0394] 2. Software:

[0395] Speech recognition engine: Google Cloud Speech-To-Text API

[0396] Emotion recognition engine: AWS Comprehend

[0397] Translation engine: Google Cloud Translation API

[0398] Speech synthesis engine: Google Cloud Text-To-Speech

[0399] Process flow and concrete examples

[0400] 1. User Action:

[0401] Users talk to their smartphones or robots, for example, a factory worker might say, "Put the new part here."

[0402] 2. Collection and Transmission of Audio Data:

[0403] The user device collects the voice data and sends it to a server via the Internet, where it is transmitted stably in digital format.

[0404] 3. Speech and Emotion Recognition:

[0405] The server receives the voice data and converts it into text data using a voice recognition engine. For example, voice data in Japanese saying "Please place the new part here" is converted into text data saying "Please place the new part here." Next, an emotion recognition engine analyzes the user's emotion from the text data and voice data and recognizes it as "neutral."

[0406] 4. Translation process:

[0407] The translation engine translates the text data into English and obtains the translation result "Please place the new part here."

[0408] 5. Speech synthesis:

[0409] The speech synthesis engine uses the translated text data and emotion data to generate a neutral voice saying, "Please place the new part here."

[0410] 6. Result output:

[0411] The generated voice data is sent to the user's device and played back through the device's speaker, enabling real-time, emotional communication between factory robots and workers who speak different languages.

[0412] Prompt Sentence Examples

[0413] A concrete example of a worker giving instructions to a robot:

[0414] Worker: Instructions in Japanese: "Put the new part here."

[0415] Robot: Outputs the instruction in English: "Please place the new part here."

[0416] In this way, the multilingual automatic translation system of the present invention realizes seamless communication that takes emotions into consideration between users who speak different languages.

[0417] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0418] Step 1:

[0419] A means of collecting user-input voice data

[0420] The user speaks into the microphone of their smartphone or robot, and the voice data is collected and converted into a digital signal, which serves as input for subsequent processing.

[0421] Step 2:

[0422] A means of sending audio data to the server

[0423] The device sends the collected audio data to a server via the internet. The audio data is sent to the server using a secure communication protocol (e.g. HTTPS). The output at this stage is digital audio data sent to the server.

[0424] Step 3:

[0425] A means of performing voice recognition and converting voice data into text data

[0426] The server passes the received voice data to a speech recognition engine (for example, Google Cloud Speech-To-Text API). The speech recognition engine analyzes the voice data and converts it into text data. The output is text data in text format, such as "Hello."

[0427] Step 4:

[0428] A means of performing emotion recognition and identifying the user's emotions

[0429] The server passes the text data and the original audio data to an emotion recognition engine (e.g., AWS Comprehend). The emotion recognition engine analyzes the text and tone of the audio to identify the user's emotion (e.g., happy, neutral, angry, etc.). The output is an emotion tag (e.g., "happy").

[0430] Step 5:

[0431] A means of translating character data into a specified language

[0432] The server passes the text data to a translation engine (e.g., Google Cloud Translation API). The translation engine translates the text data into the specified language. The output is the translated text data (e.g., "Hello").

[0433] Step 6:

[0434] A means of converting translated text data into emotionally adjusted speech data

[0435] The server passes the translated text data and emotion data to a speech synthesis engine (such as Google Cloud Text-To-Speech). The speech synthesis engine converts the text data into speech data, generating speech with intonation and tone that reflects the recognized emotion. The output is speech data adjusted based on the emotion.

[0436] Step 7:

[0437] A means for outputting audio data to a user's terminal

[0438] The server then sends the generated voice data to the user's device, which then plays the received voice data and outputs the voice from its speaker. This allows the user's interlocutor (e.g., a factory robot) to receive instructions and information in a voice that reflects the appropriate emotion.

[0439] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0440] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0441] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0442] [Second embodiment]

[0443] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0444] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0445] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0446] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0447] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0448] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0449] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0450] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0451] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0452] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0453] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0454] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0455] The multilingual machine translation system of the present invention is implemented as a smartphone application. A specific embodiment of the system will be described below.

[0456] System Overview:

[0457] The system combines speech recognition, natural language processing, and speech synthesis technologies to support real-time communication between users and speakers of other languages. Users use their smartphones to input what they want to say, and the system instantly translates and outputs the speech.

[0458] Specific steps for implementation:

[0459] User Action:

[0460] After launching the app, the user sets the "language they speak," the "language they want to convert to," and the "desired voice." The user then speaks into the smartphone's microphone, for example, saying, "Hello, what's your name?" The smartphone device collects this voice data, converts it into digital format, and sends it to a server.

[0461] Server Action:

[0462] The server passes the received voice data to a voice recognition engine, which first converts the voice into text data. This text data is then passed to a natural language processing engine, which translates it into the specified language. The translated text data is then sent to a speech synthesis engine, which generates voice data based on the "desired voice." The generated voice data is then sent from the server to the device.

[0463] The resulting output:

[0464] The terminal plays back the received voice data and lets the user and the other party hear it. For example, a voice in English saying "Hello, what is your name?" is played back.

[0465] Response from the other party:

[0466] If the other person says "My name is John" in English, the smartphone device collects this speech and sends it back to the server. The server then performs speech recognition, natural language processing, and speech synthesis in the same way, and outputs the translated speech data to the user. This allows the user to hear the Japanese audio of "My name is John" on their smartphone.

[0467] Examples:

[0468] 1. User and partner initial settings

[0469] User: Launches the app and selects "Japanese" as the spoken language and "English" as the target language. Selects "Male voice" as the desired voice.

[0470] Other person: Speak in English.

[0471] 2. User utterances

[0472] User: Say "Hello, what's your name?"

[0473] Terminal: Sends audio data to the server.

[0474] Server: The speech recognition engine converts the speech into text and saves it in the log as "Hello, what is your name?". The natural language processing engine translates it into "Hello, what is your name?". The speech synthesis engine converts the translated text into speech. The speech data is sent to the device.

[0475] Device: Play the received audio, "Hello, what is your name?"

[0476] 3. The other person's response

[0477] Other person: Say, "My name is John."

[0478] Terminal: Sends the other party's voice data to the server.

[0479] Server: The speech recognition engine converts the speech into text and saves it in the log as "My name is John." The natural language processing engine translates it as "My name is John." The speech synthesis engine converts the converted text into speech. The speech data is sent to the device.

[0480] Device: Play the received audio, "My name is John."

[0481] This allows the user and the other party to communicate seamlessly in real time. The present invention achieves high translation accuracy with almost no time lag and simple operation.

[0482] The processing flow will be explained below.

[0483] Step 1:

[0484] User: Launch the app and set the "language you speak," "language to convert to," and "desired voice" on the settings screen. For example, set "Japanese" as the language you speak, "English" as the language you want to convert to, and "male voice" as the desired voice.

[0485] Step 2:

[0486] Terminal: Saves the user's settings and enters a standby state for voice input.

[0487] Step 3:

[0488] User: Begins speaking into the smartphone microphone. For example, "Hello, what's your name?"

[0489] Step 4:

[0490] Terminal: Collects the user's voice as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[0491] Step 5:

[0492] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "Hello, what's your name?". The converted text data is recorded in a log.

[0493] Step 6:

[0494] Server: Passes the text data to the natural language processing engine. The natural language processing engine translates "Hello, what is your name?" into "Hello, what is your name?" and records the translation result in a log.

[0495] Step 7:

[0496] Server: Passes the translated text data to the speech synthesis engine. The speech synthesis engine converts "Hello, what is your name?" into audio data in a "male voice." The generated audio data is prepared for transmission to the device.

[0497] Step 8:

[0498] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "Hello, what is your name?"

[0499] Step 9:

[0500] Other person: Respond in English. For example, say, "My name is John."

[0501] Step 10:

[0502] Terminal: Collects the other party's voice data as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[0503] Step 11:

[0504] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "My name is John." The converted text data is recorded in a log.

[0505] Step 12:

[0506] Server: Passes the text data to the natural language processing engine. The natural language processing engine translates "My name is John." into "My name is John." The translation result is recorded in a log.

[0507] Step 13:

[0508] Server: Passes the translated text data to the speech synthesis engine. The speech synthesis engine converts "My name is John" into audio data in a "male voice." The generated audio data is prepared for transmission to the device.

[0509] Step 14:

[0510] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "My name is John."

[0511] Example 1

[0512] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0513] Existing multilingual translation systems often experience a time lag between speech input and speech output of the translation results, which can hinder smooth real-time communication between the user and the other party. Furthermore, the low speech quality and translation accuracy make accurate communication difficult. Furthermore, the inability to generate speech in the user's desired voice makes it difficult to provide a natural conversational experience.

[0514] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0515] In this invention, the server includes means for converting voice data input by a user into character data, means for translating the character data into a specified language, and means for converting the translated character data into voice data, thereby enabling voice output in a voice desired by the user and realizing highly accurate multilingual translation in real time.

[0516] "User" refers to the person who inputs voice data and receives the translation results.

[0517] "Voice data" refers to data that has been converted into digital form from the user's spoken voice.

[0518] "Server" refers to a central control device that processes voice data, converts it to text data, translates it, and converts it back to voice data.

[0519] "Character data" refers to data that has been converted from audio data into text format.

[0520] "Translation" refers to the process of converting character data into character data in a specified different language.

[0521] The "specified language" refers to the language into which the user wants to translate the voice data.

[0522] "Voice output" refers to the process of converting translated text data into voice data and providing it to the user.

[0523] "Digital data" refers to the digital format data used when transmitting audio data to a server.

[0524] "Desired voice" refers to the voice characteristics (e.g., gender and tone of voice) that a user uses when outputting voice.

[0525] "Terminal" refers to a device (e.g., a smartphone) that allows a user to input voice data and receive and play back the voice data of the translation result.

[0526] "Log" refers to data storage that records the voice data processed by the server and the translation results.

[0527] The multilingual machine translation system of the present invention is implemented as a smartphone app. This system translates speech data spoken by a user into other languages ​​in real time and outputs the speech in the user's desired voice, thereby smoothly supporting communication between the user and speakers of other languages. Specific embodiments of this system are described in detail below.

[0528] Overall system picture

[0529] This system combines speech recognition, natural language processing, and speech synthesis technologies, allowing users to input what they want to say by voice using their smartphone, and the system instantly translates and outputs the speech.

[0530] Hardware and software used

[0531] Hardware

[0532] Smartphone (terminal): Used for user voice input and voice output

[0533] Server: A central control unit used to process voice data.

[0534] software

[0535] Speech recognition engine: Google Cloud Speech-to-Text

[0536] Natural language processing engine: Google Cloud Translation

[0537] Speech synthesis engine: Google Cloud Text-to-Speech

[0538] Specific processing of the system

[0539] 1. User Initial Settings

[0540] The user starts the smartphone app and sets the "language they speak," the "language they want to convert to," and the "desired voice." These settings can be made on the app's settings screen.

[0541] 2. Audio input and digital conversion

[0542] Based on the user's settings, when the user speaks into the smartphone's microphone, the device converts the voice data into digital format and sends it to the server.

[0543] 3. Processing of audio data by the server

[0544] The server processes the received audio data as follows:

[0545] Speech recognition: Use the Google Cloud Speech-to-Text API to convert voice data into text data.

[0546] Natural Language Processing: Translates text data into a specified language using the Google Cloud Translation API.

[0547] Speech synthesis: Using the Google Cloud Text-to-Speech API, the translated text data is converted into voice data based on the "voice of desire."

[0548] 4. Data transmission and audio output

[0549] The server sends the generated voice data to the smartphone, which then plays it back, allowing the user to hear the translated voice.

[0550] Examples of specific examples and prompts

[0551] Specific examples

[0552] User: Launches the smartphone app, sets "Japanese" as the spoken language, "English" as the target language, and selects "male voice" as the desired voice.

[0553] User: Say "Hello, what's your name?"

[0554] Terminal: Sends audio data to the server.

[0555] Server: (Speech recognition) Convert "Hello, what's your name?" into text data.

[0556] Server: (Natural Language Processing) Translates to "Hello, what is your name?"

[0557] Server: (Speech synthesis) Generates English speech in a male voice.

[0558] Server: Sends audio data to the terminal.

[0559] Device: Play "Hello, what is your name?"

[0560] Other person: "My name is John."

[0561] Device: Sends audio to the server.

[0562] Server: (Speech recognition) Convert "My name is John." into text data.

[0563] Server: (Natural Language Processing) Translates to "My name is John."

[0564] Server: (Speech synthesis) Generates Japanese speech.

[0565] Server: Sends audio data to the terminal.

[0566] Device: Play "My name is John."

[0567] Prompt Sentence Examples

[0568] "Please translate the following phrase and play it aloud: Hello, what's your name?"

[0569] Please translate Japanese into English.

[0570] Through the above-described processing, the present invention realizes real-time, highly accurate multilingual translation and supports smooth communication between users and speakers of other languages.

[0571] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0572] Step 1:

[0573] The user launches the smartphone app and sets the "language they speak," the "language they want to translate into," and the "desired voice." This allows the user to specify which language the system will speak and which language it will translate into. For example, they can set the translation from Japanese to English and output it in a male voice.

[0574] Step 2:

[0575] The user speaks into the device's microphone, generating specific input voice data such as "Hello, what's your name?" The device converts this voice into digital data. Here, the input is an analog voice signal, and the output is digital voice data.

[0576] Step 3:

[0577] The terminal sends digital audio data to the server. This communication is carried out over a network. The input from the terminal is digital audio data, and the output to the server is also digital audio data.

[0578] Step 4:

[0579] The server converts the received voice data into text data using the Google Cloud Speech-to-Text API. The input is digital voice data, which is analyzed and the output is the text data "Hello, what's your name?"

[0580] Step 5:

[0581] The server translates the text data into the specified language using the Google Cloud Translation API. The input is Japanese text data, and the output is English text data "Hello, what is your name?". At this time, the server recognizes the language of the text and converts it into the specified language.

[0582] Step 6:

[0583] The server uses the Google Cloud Text-to-Speech API to convert the translated text data into audio data based on the desired voice. The input is the English text data "Hello, what is your name?", and the output is audio data generated in a male voice.

[0584] Step 7:

[0585] The server sends the generated voice data to the terminal. The input is the generated voice data, and the output is data transfer to the terminal. This is also done over the network.

[0586] Step 8:

[0587] The terminal plays the received voice data and lets the user and the other party listen. The terminal's input is the received voice data, and the output is the voice "Hello, what is your name?" played through the speaker.

[0588] Step 9:

[0589] The other person speaks in English into the device's microphone, "My name is John." This voice data is generated. The other person's input is an analog voice signal, and the device's output is digital voice data.

[0590] Step 10:

[0591] The terminal transmits the other party's digital voice data to the server. The input is the other party's digital voice data, and the output is the data transmitted to the server.

[0592] Step 11:

[0593] The server converts the received voice data back into text data using the Google Cloud Speech-to-Text API. The input is the other person's digital voice data, and the output is the text data "My name is John."

[0594] Step 12:

[0595] The server uses the Google Cloud Translation API to translate the text data into the specified language. The input is English text data, and the output is Japanese text data: "My name is John."

[0596] Step 13:

[0597] The server uses the Google Cloud Text-to-Speech API to convert the translated text data into audio data. The input is the Japanese text data "My name is John." and the output is audio data.

[0598] Step 14:

[0599] The server sends the generated voice data to the terminal. The input is the generated voice data, and the output is the data transfer to the terminal.

[0600] Step 15:

[0601] The terminal plays the received voice data and lets the user hear it. The input is the received voice data, and the output is the voice "My name is John" played through the speaker.

[0602] This allows for seamless real-time communication between users and speakers of other languages.

[0603] (Application example 1)

[0604] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0605] In conventional food delivery services, when the delivery person and the customer speak different languages, communication problems arise, making it difficult to deliver smoothly. This makes it difficult to immediately respond to the exact delivery location or special requests of the customer, and there is a need to improve the quality of the service.

[0606] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0607] In this invention, the server includes means for collecting voice data input by a user, means for transmitting the voice data to the server, means for the server to convert the voice data into character data, means for the server to translate the character data into a specified language, means for the server to convert the translated character data into voice data, means for outputting the translated voice data to the user, means for synthesizing the translation of the voice data into a specified language, and means for supporting communication between food delivery personnel and customers, thereby enabling smooth real-time communication between delivery personnel and customers who speak different languages.

[0608] "User-input voice data" refers to data that is collected by a device as an electrical signal when a user speaks through an input device such as a microphone.

[0609] "Means for transmitting to a server" refers to the general technology and protocols for transferring collected voice data to a server via a communications network.

[0610] "Means for converting into text data" refers to a process or engine that uses speech recognition technology to convert collected voice data into corresponding text data.

[0611] "Means for translating into a specified language" refers to technology for converting text data into a different language using a natural language processing engine.

[0612] "Means for converting into voice data" refers to a process for converting the translated text data into voice data using voice synthesis technology that synthesizes natural human speech from text.

[0613] "Means for outputting translated audio data to the user" refers to technology for playing the audio data through an output device such as a speaker.

[0614] "Means for supporting communication between food delivery personnel and customers" refers to all technologies that utilize a multilingual automatic translation system to enable real-time communication between delivery personnel and customers who speak different languages.

[0615] This invention is a multilingual automatic translation system that supports real-time communication between a user and people who speak different languages, and has an embodiment specialized for food delivery services. Specific processing and each means of this embodiment will be described below.

[0616] System Overview:

[0617] The system provides real-time multilingual translation between delivery personnel and customers, facilitating smooth communication. The system is realized by combining speech recognition, natural language processing, and speech synthesis technologies. The main hardware and software used include smartphones, servers, speech recognition engines (e.g., Google Speech-to-Text API), natural language processing engines (e.g., Google Translate API), and speech synthesis engines (e.g., gTTS).

[0618] Program processing:

[0619] Voice collection and recognition:

[0620] The device (smartphone) collects Japanese voice data input by the user. The device uses a built-in microphone to record the user's voice and sends it to the server as digital voice data. The server then uses the Google Speech-to-Text API to convert this voice data into text data.

[0621] Text translation:

[0622] The server translates the converted text data into a specified language, for example, English, using the Google Translate API. The translated text data is temporarily stored on the server.

[0623] Text-to-Speech:

[0624] The translated text data is converted into audio data using gTTS (Google Text-to-Speech), which is then sent back to the device and played through the speaker.

[0625] Examples:

[0626] User voice input:

[0627] For example, a delivery person might say in Japanese, "May I confirm your address?" The device collects this voice and sends it to the server.

[0628] Server process:

[0629] The speech recognition engine converts the speech into text and saves it in the log as "May I confirm your address?"

[0630] A natural language processing engine translates this text into English as "Can I confirm the address?"

[0631] A speech synthesis engine converts the translated text into speech.

[0632] Audio Output:

[0633] The device plays a translated English audio message that the delivery person can instantly understand in the customer's language, facilitating the confirmation of the appropriate delivery location.

[0634] Examples of prompts:

[0635] A user launches the app, sets "Japanese" as the primary language and "English" as the target language, and then says "May I confirm your address?" in Japanese. The app recognizes this, translates it into English, and plays it back aloud.

[0636] In this way, the barrier of multilingual communication in food delivery services is resolved, enabling smooth communication between delivery personnel and customers.

[0637] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0638] Step 1:

[0639] The user launches the app and provides voice input. The user launches the app on their smartphone and sets "Japanese" as the main language and "English" as the translation language. They then speak in Japanese, for example, "May I confirm your address?" The device's microphone collects this voice and processes it as digital voice data. The input in this step is the user's voice, and the output is digital voice data.

[0640] Step 2:

[0641] The device sends the collected audio data to the server. The device uploads this digital audio data to the server via an internet connection. The input in this step is the digital audio data, and the output is the audio data sent to the server.

[0642] Step 3:

[0643] The server uses a speech recognition engine to convert the transmitted voice data into text data. Specifically, the server calls the Google Speech-to-Text API to convert the voice data into text format. The input in this step is voice data, and the output is text data such as "May I confirm your address?"

[0644] Step 4:

[0645] The server passes the converted text data to a natural language processing engine and translates it into the specified language. Specifically, the server uses the Google Translate API to convert the text data "May I confirm the address?" into the English text "Can I confirm the address?". The input in this step is the text data, and the output is the translated text data.

[0646] Step 5:

[0647] The server uses a speech synthesis engine to convert the translated text data into audio data. Specifically, the server uses gTTS (Google Text-to-Speech) to convert the translated English text into an audio file. The input in this step is the translated text data, and the output is audio data.

[0648] Step 6:

[0649] The server sends the generated voice data to the terminal. The server uploads the generated voice data to the terminal via an internet connection. The input in this step is the voice data, and the output is the voice data sent to the terminal.

[0650] Step 7:

[0651] The device plays the received audio data. Specifically, it uses the device's speaker to play the translated English audio. By asking the user, "Can I confirm the address?", the delivery person can immediately understand it in the customer's language. The input in this step is audio data, and the output is audio that the user can hear.

[0652] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0653] The multilingual machine translation system of the present invention combines speech recognition, natural language processing, speech synthesis, and emotion recognition technologies to support real-time communication between users and speakers of other languages. Users input what they want to say using their smartphone, and the system instantly translates and outputs voice based on the emotion.

[0654] System Overview:

[0655] This system collects the user's voice data, performs speech recognition and emotion recognition on the server, and translates it using natural language processing. Furthermore, it synthesizes speech based on the emotion recognition results and outputs speech that reflects the appropriate emotion to the user. This not only overcomes language barriers, but also makes it possible to accurately convey emotions.

[0656] Specific steps for implementation:

[0657] User Action:

[0658] The user launches the app and sets the "language they speak," "language to convert to," "desired voice," and "emotion recognition settings" on the settings screen. The user then speaks into the smartphone's microphone, saying, for example, "Hello, what's your name?"

[0659] Audio data collection:

[0660] The terminal collects the user's voice data as a digital signal, converts it into voice data, and transmits it to a server via the Internet.

[0661] Speech and emotion recognition:

[0662] The server passes the received voice data to a voice recognition engine and an emotion engine. The voice recognition engine converts the voice data into text data, and the emotion engine recognizes the user's emotion (e.g., joy, sadness, anger, etc.).

[0663] Translation process:

[0664] The recognized text data is passed to a natural language processing engine and translated into the specified language. For example, "Hello, what is your name?" is translated into "Hello, what is your name?"

[0665] Emotion-based speech synthesis:

[0666] The server sends the translated text data and the recognized emotion data to a speech synthesis engine. The speech synthesis engine adjusts the translation result based on the emotion and generates speech data that reflects the appropriate emotion. For example, if the user is happy, the tone of the voice will also be adjusted to express happiness.

[0667] The resulting output:

[0668] The device receives the voice data from the server and plays it back through the smartphone speaker. The voice that is played back is "Hello, what is your name?", and reflects the user's emotions.

[0669] Response from the other party:

[0670] The other person responds in English, for example, "My name is John."

[0671] Processing the other party's voice data:

[0672] The device collects the other party's voice data and sends it back to the server. The server then performs speech recognition, natural language processing, and speech synthesis in the same way, and outputs the translated and emotion-reflected voice data to the user. For example, a voice saying "My name is John" is played back with the appropriate emotion.

[0673] Examples:

[0674] 1. User and partner initial settings

[0675] User: Launch the app and set "Japanese" as the speaking language, "English" as the target language, and "Male Voice" as the desired voice. Set emotion recognition to "On."

[0676] Other person: Speak in English.

[0677] 2. User utterances

[0678] User: Say "Hello, what's your name?"

[0679] Terminal: Sends audio data to the server.

[0680] Server: The speech recognition engine converts the speech into text data. It saves the message in the log as "Hello, what is your name?" The emotion engine recognizes "joy." The natural language processing engine translates it into "Hello, what is your name?" The speech synthesis engine converts it into English speech that expresses "joy." The speech data is sent to the device.

[0681] Device: Play the received audio. "Hello, what is your name?"

[0682] 3. The other person's response

[0683] Other person: Say, "My name is John."

[0684] Terminal: Sends the other party's voice data to the server.

[0685] Server: The speech recognition engine converts the speech into text data. It saves the text in the log as "My name is John." The emotion engine recognizes "neutral." The natural language processing engine translates it to "My name is John." The speech synthesis engine converts it into Japanese speech that includes "neutral." The speech data is sent to the device.

[0686] Device: Play the received audio: "My name is John."

[0687] This allows users and their partners to communicate seamlessly in real time, reflecting their emotions. This invention is a groundbreaking system in that it goes beyond simple language translation and also enables the transmission of emotions.

[0688] The processing flow will be explained below.

[0689] Step 1:

[0690] User: Launch the app and set the "language you speak," "language you want to convert to," "desired voice," and "emotion recognition settings" on the settings screen. For example, set "Japanese" as the language you speak, "English" as the language you want to convert to, "male voice" as the desired voice, and turn on "emotion recognition."

[0691] Step 2:

[0692] Terminal: Saves the user's settings and enters a standby state for voice input.

[0693] Step 3:

[0694] User: Begins speaking into the smartphone microphone. For example, "Hello, what's your name?"

[0695] Step 4:

[0696] Terminal: Collects the user's voice as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[0697] Step 5:

[0698] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "Hello, what's your name?". The converted text data is recorded in a log.

[0699] Step 6:

[0700] Server: Passes the text data to the natural language processing engine and translates it into the specified language. The natural language processing engine translates "Hello, what is your name?" into "Hello, what is your name?". The translation result is recorded in a log.

[0701] Step 7:

[0702] Server: Passes the voice data to the emotion engine and analyzes the user's emotion. The emotion engine generates an emotion tag from the voice, recognizing "joy" in this example, and records the result in a log.

[0703] Step 8:

[0704] Server: Passes the translated text data and emotion tag to the speech synthesis engine. The speech synthesis engine converts "Hello, what is your name?" into voice data in a "male voice" that reflects the emotion of "joy." The generated voice data is then prepared for transmission to the device.

[0705] Step 9:

[0706] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "Hello, what is your name?", with a tone that expresses "joy."

[0707] Step 10:

[0708] Other person: Respond in English. For example, say, "My name is John."

[0709] Step 11:

[0710] Terminal: Collects the other party's voice data again as a digital signal and sends it to a server via the Internet.

[0711] Step 12:

[0712] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "My name is John." The converted text data is recorded in a log.

[0713] Step 13:

[0714] Server: Passes the text data to the natural language processing engine and translates it into the specified language. The natural language processing engine translates "My name is John." to "My name is John." The translation result is recorded in a log.

[0715] Step 14:

[0716] Server: Passes the voice data to the emotion engine and analyzes the other person's emotion. The emotion engine generates an emotion tag from the voice, recognizing "neutral" in this example. Records the result in a log.

[0717] Step 15:

[0718] Server: Passes the translated text data and emotion tag to the speech synthesis engine. The speech synthesis engine converts "My name is John" into voice data in a "male voice" that reflects a "neutral" emotion. The generated voice data is then prepared for transmission to the device.

[0719] Step 16:

[0720] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "My name is John," in a tone that expresses "neutrality."

[0721] This step enables users and other parties to overcome language barriers and achieve real-time communication that accurately conveys emotions.

[0722] Example 2

[0723] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0724] Conventional multilingual translation systems simply translate languages ​​and have difficulty conveying emotional nuances. As a result, they are unable to accurately convey emotions, which play an important role in communication, and are therefore less convenient for actual dialogue. Furthermore, they have the problem of being difficult to translate and reflect emotions in real time, which hinders smooth communication.

[0725] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0726] In this invention, the server includes a means for converting voice data into text data, a means for recognizing emotions in the text data, and a means for converting the translated text data into voice data that reflects the emotions, thereby enabling accurate conveyance of emotions and smooth real-time multilingual communication.

[0727] "Audio data" refers to sound information input by a user through the microphone of a terminal, converted into a digital signal.

[0728] A "server" is a central processing unit that receives voice data and performs various processes such as preprocessing, voice recognition, emotion recognition, translation, and voice synthesis.

[0729] "Character data" is data in text format that has been converted from voice data by a voice recognition engine.

[0730] "Translation" is the process of converting character data into a specified language.

[0731] "Emotion recognition" is a technology that analyzes and identifies emotions (joy, surprise, sadness, etc.) from a user's voice data.

[0732] "Speech synthesis" is a technology that artificially generates natural-sounding speech based on character data and emotion recognition results.

[0733] A "terminal" is a digital device used by a user, such as a smartphone or tablet.

[0734] The multilingual machine translation system of the present invention combines speech recognition, natural language processing, speech synthesis, and emotion recognition technologies to support real-time communication between users and speakers of other languages. This system collects speech data provided by users, translates it through processing on the server, and generates and outputs speech that reflects appropriate emotions.

[0735] Hardware and Software Configuration

[0736] The main components of the system are:

[0737] 1. Device (smartphone or tablet):

[0738] A microphone that collects audio data

[0739] A function that converts audio data into a digital signal

[0740] Ability to send data to a server via the Internet

[0741] A speaker that plays back audio data received from the server

[0742] 2. Server:

[0743] Speech recognition engine (e.g., Google Cloud Speech-to-Text): converts voice data into text data

[0744] Emotion recognition engine (e.g., IBM Watson Tone Analyzer): Analyzes emotions from voice data

[0745] Natural language processing engine (e.g., Google Cloud Translation API): Translates text data into a specified language

[0746] Speech synthesis engine (e.g., Amazon Polly): Generates speech based on translated text data and emotion data.

[0747] Examples of the invention

[0748] Example 1:

[0749] The user launches the app and configures the following settings on the settings screen:

[0750] "Languages ​​I speak": "Japanese"

[0751] "Language to convert to": "English"

[0752] "Voice of Hope": "Male Voice"

[0753] "Emotion Recognition Settings": "On"

[0754] Next, the user speaks into the smartphone's microphone, saying, "Look at that dog!" The device collects this voice data as a digital signal and sends it to a server via the Internet. On the server, a voice recognition engine converts the voice data into "text data" and generates the text data "Look at that dog!" The emotion recognition engine then recognizes the user's emotion as "surprise" from the text data.

[0755] The server's natural language processing engine translates the text data into the specified language (English) and obtains the translation result "Look at that dog!". The speech synthesis engine generates English speech that reflects the appropriate emotion based on this translation result and the emotion of "surprise." The server then sends the final generated speech data to the device, and the surprised voice "Look at that dog!" is played from the device's speaker.

[0756] Example 2:

[0757] When the other person responds with "Yeah, it's really cute!", the device converts the other person's voice data into a digital signal and sends it to the server. The server uses a speech recognition engine to convert "Yeah, it's really cute!" into text data, and an emotion recognition engine recognizes the emotion "neutral." The natural language processing engine translates this to "Yeah, it's really cute!", and a speech synthesis engine generates Japanese speech that reflects a neutral tone. Finally, the generated voice data is sent to the device, and the speaker plays back the voice "Yeah, it's really cute!"

[0758] Example prompt sentence:

[0759] "The user launches the app and sets Japanese as the language they speak and English as the translation language. The user says, "Hello, what is your name?" The voice data is sent to the server, the speech recognition engine converts it to text, and the emotion engine recognizes "joy." The text data is translated by the NLP engine into "Hello, what is your name?" and the speech synthesis engine generates English speech that reflects joy. The audio is then played back on the user's device."

[0760] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0761] Step 1:

[0762] The user launches the app and enters the language they speak, the language they want to translate into, the desired voice, and emotion recognition settings on the settings screen. Specifically, they set "Japanese" as the language they speak and "English" as the target language, then select a "male voice" and turn emotion recognition "on." Once the settings are complete, the user taps the "Save Settings" button.

[0763] Input: User settings

[0764] Output: The app with the settings saved

[0765] Step 2:

[0766] After completing the setup, the user speaks into the smartphone's microphone, for example, saying, "Look at that dog!" The device converts the voice data through the microphone into a digital signal and sends this data to a server via the Internet.

[0767] Input: User's voice

[0768] Output: Audio data sent to the server

[0769] Step 3:

[0770] The server receives the voice data and passes it to a speech recognition engine. The speech recognition engine (e.g., Google Cloud Speech-to-Text) converts the voice data into text data. For example, it generates text data such as "Look at that dog!" The server then passes the text data to an emotion recognition engine (e.g., IBM Watson Tone Analyzer), which recognizes the user's emotion as "surprise" from the text data.

[0771] Input: Audio data

[0772] Output: Text data and emotion data

[0773] Step 4:

[0774] The server passes the generated text data and emotion data to a natural language processing engine. The natural language processing engine (e.g., Google Cloud Translation API) translates this text data into the specified language (English in this case). For example, "Look at that dog!" is translated to "Look at that dog!"

[0775] Input: Text data and emotion data

[0776] Output: Translated text data (English)

[0777] Step 5:

[0778] The server sends the translated text data and emotion data to a speech synthesis engine. The speech synthesis engine (e.g., Amazon Polly) generates voice data that reflects the user's emotion based on the translation results. For example, voice data that expresses surprise and says, "Look at that dog!"

[0779] Input: Translated text data and emotion data

[0780] Output: Emotion-reflecting audio data

[0781] Step 6:

[0782] The server sends the generated voice data to the device, which then plays the received voice data through the smartphone speaker, allowing the user to hear the surprised voice saying, "Look at that dog!"

[0783] Input: Audio data

[0784] Output: Audio played through the speaker

[0785] Step 7:

[0786] The other person responds to the user's speech, for example, "Yeah, it's really cute!" The device collects the other person's voice data and sends it back to the server.

[0787] Input: Other party's voice

[0788] Output: Audio data sent to the server

[0789] Step 8:

[0790] The server then passes the received voice data back to the voice recognition engine, which converts it into text data. For example, the text data generated is "Yeah, it's really cute!" The server then passes the text data to the emotion recognition engine, which analyzes the emotion and determines that it is "neutral." The translation engine translates the text data into Japanese, obtaining the translation result "Yeah, it's really cute!" Finally, the speech synthesis engine generates Japanese speech based on this translation result and the "neutral" emotion. The device then plays back the received voice data, allowing the user to hear the voice saying "Yeah, it's really cute!"

[0791] Input: Voice data of the other party

[0792] Output: Emotionally-reflected Japanese speech data

[0793] (Application example 2)

[0794] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0795] In modern factories, it is becoming increasingly common for multinational workers to work together, which can lead to communication issues between workers who speak different languages. This can lead to reduced work efficiency and potential misunderstandings and work errors. Furthermore, in addition to simple language translation, there are also issues with emotions not being conveyed when giving work instructions or reporting, which can lead to a lack of understanding of the urgency and importance of instructions. A system that can solve these problems is needed.

[0796] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0797] In this invention, the server includes means for collecting voice data input by a user, means for transmitting the voice data to the server, means for converting the voice data into character data by the server, means for translating the character data into a specified language by the server, means for converting the translated character data into voice data by the server, means for recognizing emotions from the voice data and the character data by the server, means for adjusting the translated voice data based on emotions, and means for outputting the translated voice data to a user. This enables real-time communication between workers who speak different languages ​​that takes emotions into consideration.

[0798] "Voice data" refers to digital signals that collect end-user speech in audio form.

[0799] A "server" is a central computer system that processes, converts, and translates voice data over a network.

[0800] "Character data" is information in text format converted from voice data.

[0801] "Translation" is the process of converting text data from one language into another.

[0802] "Emotion recognition" is a technology that identifies a speaker's emotions (e.g., joy, sadness, anger, etc.) from audio data and text data.

[0803] "Speech synthesis" is a technology that generates speech based on text data.

[0804] "Real-time" means that processing occurs almost immediately, with minimal time delay.

[0805] "Modification" is the process of changing the tone and intonation of speech data based on perceived emotion.

[0806] "Log" means data storage for recording and saving processed data.

[0807] The present invention is a multilingual automatic translation system that translates what a user says into other languages ​​in real time and outputs the translated content while reflecting the user's emotions. Specific embodiments for carrying out the present invention will be described below.

[0808] Program Overview

[0809] The server is composed of multiple components, including a speech recognition engine, emotion recognition engine, translation engine, and speech synthesis engine. The user's device (for example, a smartphone or robot) collects voice data and sends it to the server. When the server receives the voice data, it performs speech recognition and converts it into text data. It then performs emotion recognition to identify the user's emotion. The translation engine translates the text data into the specified language, and the speech synthesis engine generates speech based on the translated text data and emotion data. Finally, the generated voice data is sent to the user's device and played back from the device.

[0810] Hardware and software used

[0811] 1. Hardware:

[0812] User terminal: devices such as smartphones and robots

[0813] Server: High-performance computing server

[0814] 2. Software:

[0815] Speech recognition engine: Google Cloud Speech-To-Text API

[0816] Emotion recognition engine: AWS Comprehend

[0817] Translation engine: Google Cloud Translation API

[0818] Speech synthesis engine: Google Cloud Text-To-Speech

[0819] Process flow and concrete examples

[0820] 1. User Action:

[0821] Users talk to their smartphones or robots, for example, a factory worker might say, "Put the new part here."

[0822] 2. Collection and Transmission of Audio Data:

[0823] The user device collects the voice data and sends it to a server via the Internet, where it is transmitted stably in digital format.

[0824] 3. Speech and Emotion Recognition:

[0825] The server receives the voice data and converts it into text data using a voice recognition engine. For example, voice data in Japanese saying "Please place the new part here" is converted into text data saying "Please place the new part here." Next, an emotion recognition engine analyzes the user's emotion from the text data and voice data and recognizes it as "neutral."

[0826] 4. Translation process:

[0827] The translation engine translates the text data into English and obtains the translation result "Please place the new part here."

[0828] 5. Speech synthesis:

[0829] The speech synthesis engine uses the translated text data and emotion data to generate a neutral voice saying, "Please place the new part here."

[0830] 6. Result output:

[0831] The generated voice data is sent to the user's device and played back through the device's speaker, enabling real-time, emotional communication between factory robots and workers who speak different languages.

[0832] Prompt Sentence Examples

[0833] A concrete example of a worker giving instructions to a robot:

[0834] Worker: Instructions in Japanese: "Put the new part here."

[0835] Robot: Outputs the instruction in English: "Please place the new part here."

[0836] In this way, the multilingual automatic translation system of the present invention realizes seamless communication that takes emotions into consideration between users who speak different languages.

[0837] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0838] Step 1:

[0839] A means of collecting user-input voice data

[0840] The user speaks into the microphone of their smartphone or robot, and the voice data is collected and converted into a digital signal, which serves as input for subsequent processing.

[0841] Step 2:

[0842] A means of sending audio data to the server

[0843] The device sends the collected audio data to a server via the internet. The audio data is sent to the server using a secure communication protocol (e.g. HTTPS). The output at this stage is digital audio data sent to the server.

[0844] Step 3:

[0845] A means of performing voice recognition and converting voice data into text data

[0846] The server passes the received voice data to a speech recognition engine (for example, Google Cloud Speech-To-Text API). The speech recognition engine analyzes the voice data and converts it into text data. The output is text data in text format, such as "Hello."

[0847] Step 4:

[0848] A means of performing emotion recognition and identifying the user's emotions

[0849] The server passes the text data and the original audio data to an emotion recognition engine (e.g., AWS Comprehend). The emotion recognition engine analyzes the text and tone of the audio to identify the user's emotion (e.g., happy, neutral, angry, etc.). The output is an emotion tag (e.g., "happy").

[0850] Step 5:

[0851] A means of translating character data into a specified language

[0852] The server passes the text data to a translation engine (e.g., Google Cloud Translation API). The translation engine translates the text data into the specified language. The output is the translated text data (e.g., "Hello").

[0853] Step 6:

[0854] A means of converting translated text data into emotionally adjusted speech data

[0855] The server passes the translated text data and emotion data to a speech synthesis engine (such as Google Cloud Text-To-Speech). The speech synthesis engine converts the text data into speech data, generating speech with intonation and tone that reflects the recognized emotion. The output is speech data adjusted based on the emotion.

[0856] Step 7:

[0857] A means for outputting audio data to a user's terminal

[0858] The server then sends the generated voice data to the user's device, which then plays the received voice data and outputs the voice from its speaker. This allows the user's interlocutor (e.g., a factory robot) to receive instructions and information in a voice that reflects the appropriate emotion.

[0859] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0860] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0861] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0862] [Third embodiment]

[0863] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0864] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0865] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0866] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0867] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0868] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0869] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0870] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0871] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0872] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0873] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0874] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0875] The multilingual machine translation system of the present invention is implemented as a smartphone application. A specific embodiment of the system will be described below.

[0876] System Overview:

[0877] The system combines speech recognition, natural language processing, and speech synthesis technologies to support real-time communication between users and speakers of other languages. Users use their smartphones to input what they want to say, and the system instantly translates and outputs the speech.

[0878] Specific steps for implementation:

[0879] User Action:

[0880] After launching the app, the user sets the "language they speak," the "language they want to convert to," and the "desired voice." The user then speaks into the smartphone's microphone, for example, saying, "Hello, what's your name?" The smartphone device collects this voice data, converts it into digital format, and sends it to a server.

[0881] Server Action:

[0882] The server passes the received voice data to a voice recognition engine, which first converts the voice into text data. This text data is then passed to a natural language processing engine, which translates it into the specified language. The translated text data is then sent to a speech synthesis engine, which generates voice data based on the "desired voice." The generated voice data is then sent from the server to the device.

[0883] The resulting output:

[0884] The terminal plays back the received voice data and lets the user and the other party hear it. For example, a voice in English saying "Hello, what is your name?" is played back.

[0885] Response from the other party:

[0886] If the other person says "My name is John" in English, the smartphone device collects this speech and sends it back to the server. The server then performs speech recognition, natural language processing, and speech synthesis in the same way, and outputs the translated speech data to the user. This allows the user to hear the Japanese audio of "My name is John" on their smartphone.

[0887] Examples:

[0888] 1. User and partner initial settings

[0889] User: Launches the app and selects "Japanese" as the spoken language and "English" as the target language. Selects "Male voice" as the desired voice.

[0890] Other person: Speak in English.

[0891] 2. User utterances

[0892] User: Say "Hello, what's your name?"

[0893] Terminal: Sends audio data to the server.

[0894] Server: The speech recognition engine converts the speech into text and saves it in the log as "Hello, what is your name?". The natural language processing engine translates it into "Hello, what is your name?". The speech synthesis engine converts the translated text into speech. The speech data is sent to the device.

[0895] Device: Play the received audio, "Hello, what is your name?"

[0896] 3. The other person's response

[0897] Other person: Say, "My name is John."

[0898] Terminal: Sends the other party's voice data to the server.

[0899] Server: The speech recognition engine converts the speech into text and saves it in the log as "My name is John." The natural language processing engine translates it as "My name is John." The speech synthesis engine converts the converted text into speech. The speech data is sent to the device.

[0900] Device: Play the received audio, "My name is John."

[0901] This allows the user and the other party to communicate seamlessly in real time. The present invention achieves high translation accuracy with almost no time lag and simple operation.

[0902] The processing flow will be explained below.

[0903] Step 1:

[0904] User: Launch the app and set the "language you speak," "language to convert to," and "desired voice" on the settings screen. For example, set "Japanese" as the language you speak, "English" as the language you want to convert to, and "male voice" as the desired voice.

[0905] Step 2:

[0906] Terminal: Saves the user's settings and enters a standby state for voice input.

[0907] Step 3:

[0908] User: Begins speaking into the smartphone microphone. For example, "Hello, what's your name?"

[0909] Step 4:

[0910] Terminal: Collects the user's voice as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[0911] Step 5:

[0912] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "Hello, what's your name?". The converted text data is recorded in a log.

[0913] Step 6:

[0914] Server: Passes the text data to the natural language processing engine. The natural language processing engine translates "Hello, what is your name?" into "Hello, what is your name?" and records the translation result in a log.

[0915] Step 7:

[0916] Server: Passes the translated text data to the speech synthesis engine. The speech synthesis engine converts "Hello, what is your name?" into audio data in a "male voice." The generated audio data is prepared for transmission to the device.

[0917] Step 8:

[0918] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "Hello, what is your name?"

[0919] Step 9:

[0920] Other person: Respond in English. For example, say, "My name is John."

[0921] Step 10:

[0922] Terminal: Collects the other party's voice data as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[0923] Step 11:

[0924] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "My name is John." The converted text data is recorded in a log.

[0925] Step 12:

[0926] Server: Passes the text data to the natural language processing engine. The natural language processing engine translates "My name is John." into "My name is John." The translation result is recorded in a log.

[0927] Step 13:

[0928] Server: Passes the translated text data to the speech synthesis engine. The speech synthesis engine converts "My name is John" into audio data in a "male voice." The generated audio data is prepared for transmission to the device.

[0929] Step 14:

[0930] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "My name is John."

[0931] Example 1

[0932] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0933] Existing multilingual translation systems often experience a time lag between speech input and speech output of the translation results, which can hinder smooth real-time communication between the user and the other party. Furthermore, the low speech quality and translation accuracy make accurate communication difficult. Furthermore, the inability to generate speech in the user's desired voice makes it difficult to provide a natural conversational experience.

[0934] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0935] In this invention, the server includes means for converting voice data input by a user into character data, means for translating the character data into a specified language, and means for converting the translated character data into voice data, thereby enabling voice output in a voice desired by the user and realizing highly accurate multilingual translation in real time.

[0936] "User" refers to the person who inputs voice data and receives the translation results.

[0937] "Voice data" refers to data that has been converted into digital form from the user's spoken voice.

[0938] "Server" refers to a central control device that processes voice data, converts it to text data, translates it, and converts it back to voice data.

[0939] "Character data" refers to data that has been converted from audio data into text format.

[0940] "Translation" refers to the process of converting character data into character data in a specified different language.

[0941] The "specified language" refers to the language into which the user wants to translate the voice data.

[0942] "Voice output" refers to the process of converting translated text data into voice data and providing it to the user.

[0943] "Digital data" refers to the digital format data used when transmitting audio data to a server.

[0944] "Desired voice" refers to the voice characteristics (e.g., gender and tone of voice) that a user uses when outputting voice.

[0945] "Terminal" refers to a device (e.g., a smartphone) that allows a user to input voice data and receive and play back the voice data of the translation result.

[0946] "Log" refers to data storage that records the voice data processed by the server and the translation results.

[0947] The multilingual machine translation system of the present invention is implemented as a smartphone app. This system translates speech data spoken by a user into other languages ​​in real time and outputs the speech in the user's desired voice, thereby smoothly supporting communication between the user and speakers of other languages. Specific embodiments of this system are described in detail below.

[0948] Overall system picture

[0949] This system combines speech recognition, natural language processing, and speech synthesis technologies, allowing users to input what they want to say by voice using their smartphone, and the system instantly translates and outputs the speech.

[0950] Hardware and software used

[0951] Hardware

[0952] Smartphone (terminal): Used for user voice input and voice output

[0953] Server: A central control unit used to process voice data.

[0954] software

[0955] Speech recognition engine: Google Cloud Speech-to-Text

[0956] Natural language processing engine: Google Cloud Translation

[0957] Speech synthesis engine: Google Cloud Text-to-Speech

[0958] Specific processing of the system

[0959] 1. User Initial Settings

[0960] The user starts the smartphone app and sets the "language they speak," the "language they want to convert to," and the "desired voice." These settings can be made on the app's settings screen.

[0961] 2. Audio input and digital conversion

[0962] Based on the user's settings, when the user speaks into the smartphone's microphone, the device converts the voice data into digital format and sends it to the server.

[0963] 3. Processing of audio data by the server

[0964] The server processes the received audio data as follows:

[0965] Speech recognition: Use the Google Cloud Speech-to-Text API to convert voice data into text data.

[0966] Natural Language Processing: Translates text data into a specified language using the Google Cloud Translation API.

[0967] Speech synthesis: Using the Google Cloud Text-to-Speech API, the translated text data is converted into voice data based on the "voice of desire."

[0968] 4. Data transmission and audio output

[0969] The server sends the generated voice data to the smartphone, which then plays it back, allowing the user to hear the translated voice.

[0970] Examples of specific examples and prompts

[0971] Specific examples

[0972] User: Launches the smartphone app, sets "Japanese" as the spoken language, "English" as the target language, and selects "male voice" as the desired voice.

[0973] User: Say "Hello, what's your name?"

[0974] Terminal: Sends audio data to the server.

[0975] Server: (Speech recognition) Convert "Hello, what's your name?" into text data.

[0976] Server: (Natural Language Processing) Translates to "Hello, what is your name?"

[0977] Server: (Speech synthesis) Generates English speech in a male voice.

[0978] Server: Sends audio data to the terminal.

[0979] Device: Play "Hello, what is your name?"

[0980] Other person: "My name is John."

[0981] Device: Sends audio to the server.

[0982] Server: (Speech recognition) Convert "My name is John." into text data.

[0983] Server: (Natural Language Processing) Translates to "My name is John."

[0984] Server: (Speech synthesis) Generates Japanese speech.

[0985] Server: Sends audio data to the terminal.

[0986] Device: Play "My name is John."

[0987] Prompt Sentence Examples

[0988] "Please translate the following phrase and play it aloud: Hello, what's your name?"

[0989] Please translate Japanese into English.

[0990] Through the above-described processing, the present invention realizes real-time, highly accurate multilingual translation and supports smooth communication between users and speakers of other languages.

[0991] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0992] Step 1:

[0993] The user launches the smartphone app and sets the "language they speak," the "language they want to translate into," and the "desired voice." This allows the user to specify which language the system will speak and which language it will translate into. For example, they can set the translation from Japanese to English and output it in a male voice.

[0994] Step 2:

[0995] The user speaks into the device's microphone, generating specific input voice data such as "Hello, what's your name?" The device converts this voice into digital data. Here, the input is an analog voice signal, and the output is digital voice data.

[0996] Step 3:

[0997] The terminal sends digital audio data to the server. This communication is carried out over a network. The input from the terminal is digital audio data, and the output to the server is also digital audio data.

[0998] Step 4:

[0999] The server converts the received voice data into text data using the Google Cloud Speech-to-Text API. The input is digital voice data, which is analyzed and the output is the text data "Hello, what's your name?"

[1000] Step 5:

[1001] The server translates the text data into the specified language using the Google Cloud Translation API. The input is Japanese text data, and the output is English text data "Hello, what is your name?". At this time, the server recognizes the language of the text and converts it into the specified language.

[1002] Step 6:

[1003] The server uses the Google Cloud Text-to-Speech API to convert the translated text data into audio data based on the desired voice. The input is the English text data "Hello, what is your name?", and the output is audio data generated in a male voice.

[1004] Step 7:

[1005] The server sends the generated voice data to the terminal. The input is the generated voice data, and the output is data transfer to the terminal. This is also done over the network.

[1006] Step 8:

[1007] The terminal plays the received voice data and lets the user and the other party listen. The terminal's input is the received voice data, and the output is the voice "Hello, what is your name?" played through the speaker.

[1008] Step 9:

[1009] The other person speaks in English into the device's microphone, "My name is John." This voice data is generated. The other person's input is an analog voice signal, and the device's output is digital voice data.

[1010] Step 10:

[1011] The terminal transmits the other party's digital voice data to the server. The input is the other party's digital voice data, and the output is the data transmitted to the server.

[1012] Step 11:

[1013] The server converts the received voice data back into text data using the Google Cloud Speech-to-Text API. The input is the other person's digital voice data, and the output is the text data "My name is John."

[1014] Step 12:

[1015] The server uses the Google Cloud Translation API to translate the text data into the specified language. The input is English text data, and the output is Japanese text data: "My name is John."

[1016] Step 13:

[1017] The server uses the Google Cloud Text-to-Speech API to convert the translated text data into audio data. The input is the Japanese text data "My name is John." and the output is audio data.

[1018] Step 14:

[1019] The server sends the generated voice data to the terminal. The input is the generated voice data, and the output is the data transfer to the terminal.

[1020] Step 15:

[1021] The terminal plays the received voice data and lets the user hear it. The input is the received voice data, and the output is the voice "My name is John" played through the speaker.

[1022] This allows for seamless real-time communication between users and speakers of other languages.

[1023] (Application example 1)

[1024] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1025] In conventional food delivery services, when the delivery person and the customer speak different languages, communication problems arise, making it difficult to deliver smoothly. This makes it difficult to immediately respond to the exact delivery location or special requests of the customer, and there is a need to improve the quality of the service.

[1026] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1027] In this invention, the server includes means for collecting voice data input by a user, means for transmitting the voice data to the server, means for the server to convert the voice data into character data, means for the server to translate the character data into a specified language, means for the server to convert the translated character data into voice data, means for outputting the translated voice data to the user, means for synthesizing the translation of the voice data into a specified language, and means for supporting communication between food delivery personnel and customers, thereby enabling smooth real-time communication between delivery personnel and customers who speak different languages.

[1028] "User-input voice data" refers to data that is collected by a device as an electrical signal when a user speaks through an input device such as a microphone.

[1029] "Means for transmitting to a server" refers to the general technology and protocols for transferring collected voice data to a server via a communications network.

[1030] "Means for converting into text data" refers to a process or engine that uses speech recognition technology to convert collected voice data into corresponding text data.

[1031] "Means for translating into a specified language" refers to technology for converting text data into a different language using a natural language processing engine.

[1032] "Means for converting into voice data" refers to a process for converting the translated text data into voice data using voice synthesis technology that synthesizes natural human speech from text.

[1033] "Means for outputting translated audio data to the user" refers to technology for playing the audio data through an output device such as a speaker.

[1034] "Means for supporting communication between food delivery personnel and customers" refers to all technologies that utilize a multilingual automatic translation system to enable real-time communication between delivery personnel and customers who speak different languages.

[1035] This invention is a multilingual automatic translation system that supports real-time communication between a user and people who speak different languages, and has an embodiment specialized for food delivery services. Specific processing and each means of this embodiment will be described below.

[1036] System Overview:

[1037] The system provides real-time multilingual translation between delivery personnel and customers, facilitating smooth communication. The system is realized by combining speech recognition, natural language processing, and speech synthesis technologies. The main hardware and software used include smartphones, servers, speech recognition engines (e.g., Google Speech-to-Text API), natural language processing engines (e.g., Google Translate API), and speech synthesis engines (e.g., gTTS).

[1038] Program processing:

[1039] Voice collection and recognition:

[1040] The device (smartphone) collects Japanese voice data input by the user. The device uses a built-in microphone to record the user's voice and sends it to the server as digital voice data. The server then uses the Google Speech-to-Text API to convert this voice data into text data.

[1041] Text translation:

[1042] The server translates the converted text data into a specified language, for example, English, using the Google Translate API. The translated text data is temporarily stored on the server.

[1043] Text-to-Speech:

[1044] The translated text data is converted into audio data using gTTS (Google Text-to-Speech), which is then sent back to the device and played through the speaker.

[1045] Examples:

[1046] User voice input:

[1047] For example, a delivery person might say in Japanese, "May I confirm your address?" The device collects this voice and sends it to the server.

[1048] Server process:

[1049] The speech recognition engine converts the speech into text and saves it in the log as "May I confirm your address?"

[1050] A natural language processing engine translates this text into English as "Can I confirm the address?"

[1051] A speech synthesis engine converts the translated text into speech.

[1052] Audio Output:

[1053] The device plays a translated English audio message that the delivery person can instantly understand in the customer's language, facilitating the confirmation of the appropriate delivery location.

[1054] Examples of prompts:

[1055] A user launches the app, sets "Japanese" as the primary language and "English" as the target language, and then says "May I confirm your address?" in Japanese. The app recognizes this, translates it into English, and plays it back aloud.

[1056] In this way, the barrier of multilingual communication in food delivery services is resolved, enabling smooth communication between delivery personnel and customers.

[1057] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1058] Step 1:

[1059] The user launches the app and provides voice input. The user launches the app on their smartphone and sets "Japanese" as the main language and "English" as the translation language. They then speak in Japanese, for example, "May I confirm your address?" The device's microphone collects this voice and processes it as digital voice data. The input in this step is the user's voice, and the output is digital voice data.

[1060] Step 2:

[1061] The device sends the collected audio data to the server. The device uploads this digital audio data to the server via an internet connection. The input in this step is the digital audio data, and the output is the audio data sent to the server.

[1062] Step 3:

[1063] The server uses a speech recognition engine to convert the transmitted voice data into text data. Specifically, the server calls the Google Speech-to-Text API to convert the voice data into text format. The input in this step is voice data, and the output is text data such as "May I confirm your address?"

[1064] Step 4:

[1065] The server passes the converted text data to a natural language processing engine and translates it into the specified language. Specifically, the server uses the Google Translate API to convert the text data "May I confirm the address?" into the English text "Can I confirm the address?". The input in this step is the text data, and the output is the translated text data.

[1066] Step 5:

[1067] The server uses a speech synthesis engine to convert the translated text data into audio data. Specifically, the server uses gTTS (Google Text-to-Speech) to convert the translated English text into an audio file. The input in this step is the translated text data, and the output is audio data.

[1068] Step 6:

[1069] The server sends the generated voice data to the terminal. The server uploads the generated voice data to the terminal via an internet connection. The input in this step is the voice data, and the output is the voice data sent to the terminal.

[1070] Step 7:

[1071] The device plays the received audio data. Specifically, it uses the device's speaker to play the translated English audio. By asking the user, "Can I confirm the address?", the delivery person can immediately understand it in the customer's language. The input in this step is audio data, and the output is audio that the user can hear.

[1072] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1073] The multilingual machine translation system of the present invention combines speech recognition, natural language processing, speech synthesis, and emotion recognition technologies to support real-time communication between users and speakers of other languages. Users input what they want to say using their smartphone, and the system instantly translates and outputs voice based on the emotion.

[1074] System Overview:

[1075] This system collects the user's voice data, performs speech recognition and emotion recognition on the server, and translates it using natural language processing. Furthermore, it synthesizes speech based on the emotion recognition results and outputs speech that reflects the appropriate emotion to the user. This not only overcomes language barriers, but also makes it possible to accurately convey emotions.

[1076] Specific steps for implementation:

[1077] User Action:

[1078] The user launches the app and sets the "language they speak," "language to convert to," "desired voice," and "emotion recognition settings" on the settings screen. The user then speaks into the smartphone's microphone, saying, for example, "Hello, what's your name?"

[1079] Audio data collection:

[1080] The terminal collects the user's voice data as a digital signal, converts it into voice data, and transmits it to a server via the Internet.

[1081] Speech and emotion recognition:

[1082] The server passes the received voice data to a voice recognition engine and an emotion engine. The voice recognition engine converts the voice data into text data, and the emotion engine recognizes the user's emotion (e.g., joy, sadness, anger, etc.).

[1083] Translation process:

[1084] The recognized text data is passed to a natural language processing engine and translated into the specified language. For example, "Hello, what is your name?" is translated into "Hello, what is your name?"

[1085] Emotion-based speech synthesis:

[1086] The server sends the translated text data and the recognized emotion data to a speech synthesis engine. The speech synthesis engine adjusts the translation result based on the emotion and generates speech data that reflects the appropriate emotion. For example, if the user is happy, the tone of the voice will also be adjusted to express happiness.

[1087] The resulting output:

[1088] The device receives the voice data from the server and plays it back through the smartphone speaker. The voice that is played back is "Hello, what is your name?", and reflects the user's emotions.

[1089] Response from the other party:

[1090] The other person responds in English, for example, "My name is John."

[1091] Processing the other party's voice data:

[1092] The device collects the other party's voice data and sends it back to the server. The server then performs speech recognition, natural language processing, and speech synthesis in the same way, and outputs the translated and emotion-reflected voice data to the user. For example, a voice saying "My name is John" is played back with the appropriate emotion.

[1093] Examples:

[1094] 1. User and partner initial settings

[1095] User: Launch the app and set "Japanese" as the speaking language, "English" as the target language, and "Male Voice" as the desired voice. Set emotion recognition to "On."

[1096] Other person: Speak in English.

[1097] 2. User utterances

[1098] User: Say "Hello, what's your name?"

[1099] Terminal: Sends audio data to the server.

[1100] Server: The speech recognition engine converts the speech into text data. It saves the message in the log as "Hello, what is your name?" The emotion engine recognizes "joy." The natural language processing engine translates it into "Hello, what is your name?" The speech synthesis engine converts it into English speech that expresses "joy." The speech data is sent to the device.

[1101] Device: Play the received audio. "Hello, what is your name?"

[1102] 3. The other person's response

[1103] Other person: Say, "My name is John."

[1104] Terminal: Sends the other party's voice data to the server.

[1105] Server: The speech recognition engine converts the speech into text data. It saves the text in the log as "My name is John." The emotion engine recognizes "neutral." The natural language processing engine translates it to "My name is John." The speech synthesis engine converts it into Japanese speech that includes "neutral." The speech data is sent to the device.

[1106] Device: Play the received audio: "My name is John."

[1107] This allows users and their partners to communicate seamlessly in real time, reflecting their emotions. This invention is a groundbreaking system in that it goes beyond simple language translation and also enables the transmission of emotions.

[1108] The processing flow will be explained below.

[1109] Step 1:

[1110] User: Launch the app and set the "language you speak," "language you want to convert to," "desired voice," and "emotion recognition settings" on the settings screen. For example, set "Japanese" as the language you speak, "English" as the language you want to convert to, "male voice" as the desired voice, and turn on "emotion recognition."

[1111] Step 2:

[1112] Terminal: Saves the user's settings and enters a standby state for voice input.

[1113] Step 3:

[1114] User: Begins speaking into the smartphone microphone. For example, "Hello, what's your name?"

[1115] Step 4:

[1116] Terminal: Collects the user's voice as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[1117] Step 5:

[1118] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "Hello, what's your name?". The converted text data is recorded in a log.

[1119] Step 6:

[1120] Server: Passes the text data to the natural language processing engine and translates it into the specified language. The natural language processing engine translates "Hello, what is your name?" into "Hello, what is your name?". The translation result is recorded in a log.

[1121] Step 7:

[1122] Server: Passes the voice data to the emotion engine and analyzes the user's emotion. The emotion engine generates an emotion tag from the voice, recognizing "joy" in this example, and records the result in a log.

[1123] Step 8:

[1124] Server: Passes the translated text data and emotion tag to the speech synthesis engine. The speech synthesis engine converts "Hello, what is your name?" into voice data in a "male voice" that reflects the emotion of "joy." The generated voice data is then prepared for transmission to the device.

[1125] Step 9:

[1126] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "Hello, what is your name?", with a tone that expresses "joy."

[1127] Step 10:

[1128] Other person: Respond in English. For example, say, "My name is John."

[1129] Step 11:

[1130] Terminal: Collects the other party's voice data again as a digital signal and sends it to a server via the Internet.

[1131] Step 12:

[1132] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "My name is John." The converted text data is recorded in a log.

[1133] Step 13:

[1134] Server: Passes the text data to the natural language processing engine and translates it into the specified language. The natural language processing engine translates "My name is John." to "My name is John." The translation result is recorded in a log.

[1135] Step 14:

[1136] Server: Passes the voice data to the emotion engine and analyzes the other person's emotion. The emotion engine generates an emotion tag from the voice, recognizing "neutral" in this example. Records the result in a log.

[1137] Step 15:

[1138] Server: Passes the translated text data and emotion tag to the speech synthesis engine. The speech synthesis engine converts "My name is John" into voice data in a "male voice" that reflects a "neutral" emotion. The generated voice data is then prepared for transmission to the device.

[1139] Step 16:

[1140] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "My name is John," in a tone that expresses "neutrality."

[1141] This step enables users and other parties to overcome language barriers and achieve real-time communication that accurately conveys emotions.

[1142] Example 2

[1143] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1144] Conventional multilingual translation systems simply translate languages ​​and have difficulty conveying emotional nuances. As a result, they are unable to accurately convey emotions, which play an important role in communication, and are therefore less convenient for actual dialogue. Furthermore, they have the problem of being difficult to translate and reflect emotions in real time, which hinders smooth communication.

[1145] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1146] In this invention, the server includes a means for converting voice data into text data, a means for recognizing emotions in the text data, and a means for converting the translated text data into voice data that reflects the emotions, thereby enabling accurate conveyance of emotions and smooth real-time multilingual communication.

[1147] "Audio data" refers to sound information input by a user through the microphone of a terminal, converted into a digital signal.

[1148] A "server" is a central processing unit that receives voice data and performs various processes such as preprocessing, voice recognition, emotion recognition, translation, and voice synthesis.

[1149] "Character data" is data in text format that has been converted from voice data by a voice recognition engine.

[1150] "Translation" is the process of converting character data into a specified language.

[1151] "Emotion recognition" is a technology that analyzes and identifies emotions (joy, surprise, sadness, etc.) from a user's voice data.

[1152] "Speech synthesis" is a technology that artificially generates natural-sounding speech based on character data and emotion recognition results.

[1153] A "terminal" is a digital device used by a user, such as a smartphone or tablet.

[1154] The multilingual machine translation system of the present invention combines speech recognition, natural language processing, speech synthesis, and emotion recognition technologies to support real-time communication between users and speakers of other languages. This system collects speech data provided by users, translates it through processing on the server, and generates and outputs speech that reflects appropriate emotions.

[1155] Hardware and Software Configuration

[1156] The main components of the system are:

[1157] 1. Device (smartphone or tablet):

[1158] A microphone that collects audio data

[1159] A function that converts audio data into a digital signal

[1160] Ability to send data to a server via the Internet

[1161] A speaker that plays back audio data received from the server

[1162] 2. Server:

[1163] Speech recognition engine (e.g., Google Cloud Speech-to-Text): converts voice data into text data

[1164] Emotion recognition engine (e.g., IBM Watson Tone Analyzer): Analyzes emotions from voice data

[1165] Natural language processing engine (e.g., Google Cloud Translation API): Translates text data into a specified language

[1166] Speech synthesis engine (e.g., Amazon Polly): Generates speech based on translated text data and emotion data.

[1167] Examples of the invention

[1168] Example 1:

[1169] The user launches the app and configures the following settings on the settings screen:

[1170] "Languages ​​I speak": "Japanese"

[1171] "Language to convert to": "English"

[1172] "Voice of Hope": "Male Voice"

[1173] "Emotion Recognition Settings": "On"

[1174] Next, the user speaks into the smartphone's microphone, saying, "Look at that dog!" The device collects this voice data as a digital signal and sends it to a server via the Internet. On the server, a voice recognition engine converts the voice data into "text data" and generates the text data "Look at that dog!" The emotion recognition engine then recognizes the user's emotion as "surprise" from the text data.

[1175] The server's natural language processing engine translates the text data into the specified language (English) and obtains the translation result "Look at that dog!". The speech synthesis engine generates English speech that reflects the appropriate emotion based on this translation result and the emotion of "surprise." The server then sends the final generated speech data to the device, and the surprised voice "Look at that dog!" is played from the device's speaker.

[1176] Example 2:

[1177] When the other person responds with "Yeah, it's really cute!", the device converts the other person's voice data into a digital signal and sends it to the server. The server uses a speech recognition engine to convert "Yeah, it's really cute!" into text data, and an emotion recognition engine recognizes the emotion "neutral." The natural language processing engine translates this to "Yeah, it's really cute!", and a speech synthesis engine generates Japanese speech that reflects a neutral tone. Finally, the generated voice data is sent to the device, and the speaker plays back the voice "Yeah, it's really cute!"

[1178] Example prompt sentence:

[1179] "The user launches the app and sets Japanese as the language they speak and English as the translation language. The user says, "Hello, what is your name?" The voice data is sent to the server, the speech recognition engine converts it to text, and the emotion engine recognizes "joy." The text data is translated by the NLP engine into "Hello, what is your name?" and the speech synthesis engine generates English speech that reflects joy. The audio is then played back on the user's device."

[1180] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1181] Step 1:

[1182] The user launches the app and enters the language they speak, the language they want to translate into, the desired voice, and emotion recognition settings on the settings screen. Specifically, they set "Japanese" as the language they speak and "English" as the target language, then select a "male voice" and turn emotion recognition "on." Once the settings are complete, the user taps the "Save Settings" button.

[1183] Input: User settings

[1184] Output: The app with the settings saved

[1185] Step 2:

[1186] After completing the setup, the user speaks into the smartphone's microphone, for example, saying, "Look at that dog!" The device converts the voice data through the microphone into a digital signal and sends this data to a server via the Internet.

[1187] Input: User's voice

[1188] Output: Audio data sent to the server

[1189] Step 3:

[1190] The server receives the voice data and passes it to a speech recognition engine. The speech recognition engine (e.g., Google Cloud Speech-to-Text) converts the voice data into text data. For example, it generates text data such as "Look at that dog!" The server then passes the text data to an emotion recognition engine (e.g., IBM Watson Tone Analyzer), which recognizes the user's emotion as "surprise" from the text data.

[1191] Input: Audio data

[1192] Output: Text data and emotion data

[1193] Step 4:

[1194] The server passes the generated text data and emotion data to a natural language processing engine. The natural language processing engine (e.g., Google Cloud Translation API) translates this text data into the specified language (English in this case). For example, "Look at that dog!" is translated to "Look at that dog!"

[1195] Input: Text data and emotion data

[1196] Output: Translated text data (English)

[1197] Step 5:

[1198] The server sends the translated text data and emotion data to a speech synthesis engine. The speech synthesis engine (e.g., Amazon Polly) generates voice data that reflects the user's emotion based on the translation results. For example, voice data that expresses surprise and says, "Look at that dog!"

[1199] Input: Translated text data and emotion data

[1200] Output: Emotion-reflecting audio data

[1201] Step 6:

[1202] The server sends the generated voice data to the device, which then plays the received voice data through the smartphone speaker, allowing the user to hear the surprised voice saying, "Look at that dog!"

[1203] Input: Audio data

[1204] Output: Audio played through the speaker

[1205] Step 7:

[1206] The other person responds to the user's speech, for example, "Yeah, it's really cute!" The device collects the other person's voice data and sends it back to the server.

[1207] Input: Other party's voice

[1208] Output: Audio data sent to the server

[1209] Step 8:

[1210] The server then passes the received voice data back to the voice recognition engine, which converts it into text data. For example, the text data generated is "Yeah, it's really cute!" The server then passes the text data to the emotion recognition engine, which analyzes the emotion and determines that it is "neutral." The translation engine translates the text data into Japanese, obtaining the translation result "Yeah, it's really cute!" Finally, the speech synthesis engine generates Japanese speech based on this translation result and the "neutral" emotion. The device then plays back the received voice data, allowing the user to hear the voice saying "Yeah, it's really cute!"

[1211] Input: Voice data of the other party

[1212] Output: Emotionally-reflected Japanese speech data

[1213] (Application example 2)

[1214] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1215] In modern factories, it is becoming increasingly common for multinational workers to work together, which can lead to communication issues between workers who speak different languages. This can lead to reduced work efficiency and potential misunderstandings and work errors. Furthermore, in addition to simple language translation, there are also issues with emotions not being conveyed when giving work instructions or reporting, which can lead to a lack of understanding of the urgency and importance of instructions. A system that can solve these problems is needed.

[1216] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1217] In this invention, the server includes means for collecting voice data input by a user, means for transmitting the voice data to the server, means for converting the voice data into character data by the server, means for translating the character data into a specified language by the server, means for converting the translated character data into voice data by the server, means for recognizing emotions from the voice data and the character data by the server, means for adjusting the translated voice data based on emotions, and means for outputting the translated voice data to a user. This enables real-time communication between workers who speak different languages ​​that takes emotions into consideration.

[1218] "Voice data" refers to digital signals that collect end-user speech in audio form.

[1219] A "server" is a central computer system that processes, converts, and translates voice data over a network.

[1220] "Character data" is information in text format converted from voice data.

[1221] "Translation" is the process of converting text data from one language into another.

[1222] "Emotion recognition" is a technology that identifies a speaker's emotions (e.g., joy, sadness, anger, etc.) from audio data and text data.

[1223] "Speech synthesis" is a technology that generates speech based on text data.

[1224] "Real-time" means that processing occurs almost immediately, with minimal time delay.

[1225] "Modification" is the process of changing the tone and intonation of speech data based on perceived emotion.

[1226] "Log" means data storage for recording and saving processed data.

[1227] The present invention is a multilingual automatic translation system that translates what a user says into other languages ​​in real time and outputs the translated content while reflecting the user's emotions. Specific embodiments for carrying out the present invention will be described below.

[1228] Program Overview

[1229] The server is composed of multiple components, including a speech recognition engine, emotion recognition engine, translation engine, and speech synthesis engine. The user's device (for example, a smartphone or robot) collects voice data and sends it to the server. When the server receives the voice data, it performs speech recognition and converts it into text data. It then performs emotion recognition to identify the user's emotion. The translation engine translates the text data into the specified language, and the speech synthesis engine generates speech based on the translated text data and emotion data. Finally, the generated voice data is sent to the user's device and played back from the device.

[1230] Hardware and software used

[1231] 1. Hardware:

[1232] User terminal: devices such as smartphones and robots

[1233] Server: High-performance computing server

[1234] 2. Software:

[1235] Speech recognition engine: Google Cloud Speech-To-Text API

[1236] Emotion recognition engine: AWS Comprehend

[1237] Translation engine: Google Cloud Translation API

[1238] Speech synthesis engine: Google Cloud Text-To-Speech

[1239] Process flow and concrete examples

[1240] 1. User Action:

[1241] Users talk to their smartphones or robots, for example, a factory worker might say, "Put the new part here."

[1242] 2. Collection and Transmission of Audio Data:

[1243] The user device collects the voice data and sends it to a server via the Internet, where it is transmitted stably in digital format.

[1244] 3. Speech and Emotion Recognition:

[1245] The server receives the voice data and converts it into text data using a voice recognition engine. For example, voice data in Japanese saying "Please place the new part here" is converted into text data saying "Please place the new part here." Next, an emotion recognition engine analyzes the user's emotion from the text data and voice data and recognizes it as "neutral."

[1246] 4. Translation process:

[1247] The translation engine translates the text data into English and obtains the translation result "Please place the new part here."

[1248] 5. Speech synthesis:

[1249] The speech synthesis engine uses the translated text data and emotion data to generate a neutral voice saying, "Please place the new part here."

[1250] 6. Result output:

[1251] The generated voice data is sent to the user's device and played back through the device's speaker, enabling real-time, emotional communication between factory robots and workers who speak different languages.

[1252] Prompt Sentence Examples

[1253] A concrete example of a worker giving instructions to a robot:

[1254] Worker: Instructions in Japanese: "Put the new part here."

[1255] Robot: Outputs the instruction in English: "Please place the new part here."

[1256] In this way, the multilingual automatic translation system of the present invention realizes seamless communication that takes emotions into consideration between users who speak different languages.

[1257] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1258] Step 1:

[1259] A means of collecting user-input voice data

[1260] The user speaks into the microphone of their smartphone or robot, and the voice data is collected and converted into a digital signal, which serves as input for subsequent processing.

[1261] Step 2:

[1262] A means of sending audio data to the server

[1263] The device sends the collected audio data to a server via the internet. The audio data is sent to the server using a secure communication protocol (e.g. HTTPS). The output at this stage is digital audio data sent to the server.

[1264] Step 3:

[1265] A means of performing voice recognition and converting voice data into text data

[1266] The server passes the received voice data to a speech recognition engine (for example, Google Cloud Speech-To-Text API). The speech recognition engine analyzes the voice data and converts it into text data. The output is text data in text format, such as "Hello."

[1267] Step 4:

[1268] A means of performing emotion recognition and identifying the user's emotions

[1269] The server passes the text data and the original audio data to an emotion recognition engine (e.g., AWS Comprehend). The emotion recognition engine analyzes the text and tone of the audio to identify the user's emotion (e.g., happy, neutral, angry, etc.). The output is an emotion tag (e.g., "happy").

[1270] Step 5:

[1271] A means of translating character data into a specified language

[1272] The server passes the text data to a translation engine (e.g., Google Cloud Translation API). The translation engine translates the text data into the specified language. The output is the translated text data (e.g., "Hello").

[1273] Step 6:

[1274] A means of converting translated text data into emotionally adjusted speech data

[1275] The server passes the translated text data and emotion data to a speech synthesis engine (such as Google Cloud Text-To-Speech). The speech synthesis engine converts the text data into speech data, generating speech with intonation and tone that reflects the recognized emotion. The output is speech data adjusted based on the emotion.

[1276] Step 7:

[1277] A means for outputting audio data to a user's terminal

[1278] The server then sends the generated voice data to the user's device, which then plays the received voice data and outputs the voice from its speaker. This allows the user's interlocutor (e.g., a factory robot) to receive instructions and information in a voice that reflects the appropriate emotion.

[1279] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1280] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1281] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1282] [Fourth embodiment]

[1283] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1284] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1285] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1286] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1287] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1288] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1289] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1290] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1291] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1292] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1293] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1294] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1295] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1296] The multilingual machine translation system of the present invention is implemented as a smartphone application. A specific embodiment of the system will be described below.

[1297] System Overview:

[1298] The system combines speech recognition, natural language processing, and speech synthesis technologies to support real-time communication between users and speakers of other languages. Users use their smartphones to input what they want to say, and the system instantly translates and outputs the speech.

[1299] Specific steps for implementation:

[1300] User Action:

[1301] After launching the app, the user sets the "language they speak," the "language they want to convert to," and the "desired voice." The user then speaks into the smartphone's microphone, for example, saying, "Hello, what's your name?" The smartphone device collects this voice data, converts it into digital format, and sends it to a server.

[1302] Server Action:

[1303] The server passes the received voice data to a voice recognition engine, which first converts the voice into text data. This text data is then passed to a natural language processing engine, which translates it into the specified language. The translated text data is then sent to a speech synthesis engine, which generates voice data based on the "desired voice." The generated voice data is then sent from the server to the device.

[1304] The resulting output:

[1305] The terminal plays back the received voice data and lets the user and the other party hear it. For example, a voice in English saying "Hello, what is your name?" is played back.

[1306] Response from the other party:

[1307] If the other person says "My name is John" in English, the smartphone device collects this speech and sends it back to the server. The server then performs speech recognition, natural language processing, and speech synthesis in the same way, and outputs the translated speech data to the user. This allows the user to hear the Japanese audio of "My name is John" on their smartphone.

[1308] Examples:

[1309] 1. User and partner initial settings

[1310] User: Launches the app and selects "Japanese" as the spoken language and "English" as the target language. Selects "Male voice" as the desired voice.

[1311] Other person: Speak in English.

[1312] 2. User utterances

[1313] User: Say "Hello, what's your name?"

[1314] Terminal: Sends audio data to the server.

[1315] Server: The speech recognition engine converts the speech into text and saves it in the log as "Hello, what is your name?". The natural language processing engine translates it into "Hello, what is your name?". The speech synthesis engine converts the translated text into speech. The speech data is sent to the device.

[1316] Device: Play the received audio, "Hello, what is your name?"

[1317] 3. The other person's response

[1318] Other person: Say, "My name is John."

[1319] Terminal: Sends the other party's voice data to the server.

[1320] Server: The speech recognition engine converts the speech into text and saves it in the log as "My name is John." The natural language processing engine translates it as "My name is John." The speech synthesis engine converts the converted text into speech. The speech data is sent to the device.

[1321] Device: Play the received audio, "My name is John."

[1322] This allows the user and the other party to communicate seamlessly in real time. The present invention achieves high translation accuracy with almost no time lag and simple operation.

[1323] The processing flow will be explained below.

[1324] Step 1:

[1325] User: Launch the app and set the "language you speak," "language to convert to," and "desired voice" on the settings screen. For example, set "Japanese" as the language you speak, "English" as the language you want to convert to, and "male voice" as the desired voice.

[1326] Step 2:

[1327] Terminal: Saves the user's settings and enters a standby state for voice input.

[1328] Step 3:

[1329] User: Begins speaking into the smartphone microphone. For example, "Hello, what's your name?"

[1330] Step 4:

[1331] Terminal: Collects the user's voice as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[1332] Step 5:

[1333] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "Hello, what's your name?". The converted text data is recorded in a log.

[1334] Step 6:

[1335] Server: Passes the text data to the natural language processing engine. The natural language processing engine translates "Hello, what is your name?" into "Hello, what is your name?" and records the translation result in a log.

[1336] Step 7:

[1337] Server: Passes the translated text data to the speech synthesis engine. The speech synthesis engine converts "Hello, what is your name?" into audio data in a "male voice." The generated audio data is prepared for transmission to the device.

[1338] Step 8:

[1339] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "Hello, what is your name?"

[1340] Step 9:

[1341] Other person: Respond in English. For example, say, "My name is John."

[1342] Step 10:

[1343] Terminal: Collects the other party's voice data as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[1344] Step 11:

[1345] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "My name is John." The converted text data is recorded in a log.

[1346] Step 12:

[1347] Server: Passes the text data to the natural language processing engine. The natural language processing engine translates "My name is John." into "My name is John." The translation result is recorded in a log.

[1348] Step 13:

[1349] Server: Passes the translated text data to the speech synthesis engine. The speech synthesis engine converts "My name is John" into audio data in a "male voice." The generated audio data is prepared for transmission to the device.

[1350] Step 14:

[1351] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "My name is John."

[1352] Example 1

[1353] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1354] Existing multilingual translation systems often experience a time lag between speech input and speech output of the translation results, which can hinder smooth real-time communication between the user and the other party. Furthermore, the low speech quality and translation accuracy make accurate communication difficult. Furthermore, the inability to generate speech in the user's desired voice makes it difficult to provide a natural conversational experience.

[1355] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1356] In this invention, the server includes means for converting voice data input by a user into character data, means for translating the character data into a specified language, and means for converting the translated character data into voice data, thereby enabling voice output in a voice desired by the user and realizing highly accurate multilingual translation in real time.

[1357] "User" refers to the person who inputs voice data and receives the translation results.

[1358] "Voice data" refers to data that has been converted into digital form from the user's spoken voice.

[1359] "Server" refers to a central control device that processes voice data, converts it to text data, translates it, and converts it back to voice data.

[1360] "Character data" refers to data that has been converted from audio data into text format.

[1361] "Translation" refers to the process of converting character data into character data in a specified different language.

[1362] The "specified language" refers to the language into which the user wants to translate the voice data.

[1363] "Voice output" refers to the process of converting translated text data into voice data and providing it to the user.

[1364] "Digital data" refers to the digital format data used when transmitting audio data to a server.

[1365] "Desired voice" refers to the voice characteristics (e.g., gender and tone of voice) that a user uses when outputting voice.

[1366] "Terminal" refers to a device (e.g., a smartphone) that allows a user to input voice data and receive and play back the voice data of the translation result.

[1367] "Log" refers to data storage that records the voice data processed by the server and the translation results.

[1368] The multilingual machine translation system of the present invention is implemented as a smartphone app. This system translates speech data spoken by a user into other languages ​​in real time and outputs the speech in the user's desired voice, thereby smoothly supporting communication between the user and speakers of other languages. Specific embodiments of this system are described in detail below.

[1369] Overall system picture

[1370] This system combines speech recognition, natural language processing, and speech synthesis technologies, allowing users to input what they want to say by voice using their smartphone, and the system instantly translates and outputs the speech.

[1371] Hardware and software used

[1372] Hardware

[1373] Smartphone (terminal): Used for user voice input and voice output

[1374] Server: A central control unit used to process voice data.

[1375] software

[1376] Speech recognition engine: Google Cloud Speech-to-Text

[1377] Natural language processing engine: Google Cloud Translation

[1378] Speech synthesis engine: Google Cloud Text-to-Speech

[1379] Specific processing of the system

[1380] 1. User Initial Settings

[1381] The user starts the smartphone app and sets the "language they speak," the "language they want to convert to," and the "desired voice." These settings can be made on the app's settings screen.

[1382] 2. Audio input and digital conversion

[1383] Based on the user's settings, when the user speaks into the smartphone's microphone, the device converts the voice data into digital format and sends it to the server.

[1384] 3. Processing of audio data by the server

[1385] The server processes the received audio data as follows:

[1386] Speech recognition: Use the Google Cloud Speech-to-Text API to convert voice data into text data.

[1387] Natural Language Processing: Translates text data into a specified language using the Google Cloud Translation API.

[1388] Speech synthesis: Using the Google Cloud Text-to-Speech API, the translated text data is converted into voice data based on the "voice of desire."

[1389] 4. Data transmission and audio output

[1390] The server sends the generated voice data to the smartphone, which then plays it back, allowing the user to hear the translated voice.

[1391] Examples of specific examples and prompts

[1392] Specific examples

[1393] User: Launches the smartphone app, sets "Japanese" as the spoken language, "English" as the target language, and selects "male voice" as the desired voice.

[1394] User: Say "Hello, what's your name?"

[1395] Terminal: Sends audio data to the server.

[1396] Server: (Speech recognition) Convert "Hello, what's your name?" into text data.

[1397] Server: (Natural Language Processing) Translates to "Hello, what is your name?"

[1398] Server: (Speech synthesis) Generates English speech in a male voice.

[1399] Server: Sends audio data to the terminal.

[1400] Device: Play "Hello, what is your name?"

[1401] Other person: "My name is John."

[1402] Device: Sends audio to the server.

[1403] Server: (Speech recognition) Convert "My name is John." into text data.

[1404] Server: (Natural Language Processing) Translates to "My name is John."

[1405] Server: (Speech synthesis) Generates Japanese speech.

[1406] Server: Sends audio data to the terminal.

[1407] Device: Play "My name is John."

[1408] Prompt Sentence Examples

[1409] "Please translate the following phrase and play it aloud: Hello, what's your name?"

[1410] Please translate Japanese into English.

[1411] Through the above-described processing, the present invention realizes real-time, highly accurate multilingual translation and supports smooth communication between users and speakers of other languages.

[1412] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1413] Step 1:

[1414] The user launches the smartphone app and sets the "language they speak," the "language they want to translate into," and the "desired voice." This allows the user to specify which language the system will speak and which language it will translate into. For example, they can set the translation from Japanese to English and output it in a male voice.

[1415] Step 2:

[1416] The user speaks into the device's microphone, generating specific input voice data such as "Hello, what's your name?" The device converts this voice into digital data. Here, the input is an analog voice signal, and the output is digital voice data.

[1417] Step 3:

[1418] The terminal sends digital audio data to the server. This communication is carried out over a network. The input from the terminal is digital audio data, and the output to the server is also digital audio data.

[1419] Step 4:

[1420] The server converts the received voice data into text data using the Google Cloud Speech-to-Text API. The input is digital voice data, which is analyzed and the output is the text data "Hello, what's your name?"

[1421] Step 5:

[1422] The server translates the text data into the specified language using the Google Cloud Translation API. The input is Japanese text data, and the output is English text data "Hello, what is your name?". At this time, the server recognizes the language of the text and converts it into the specified language.

[1423] Step 6:

[1424] The server uses the Google Cloud Text-to-Speech API to convert the translated text data into audio data based on the desired voice. The input is the English text data "Hello, what is your name?", and the output is audio data generated in a male voice.

[1425] Step 7:

[1426] The server sends the generated voice data to the terminal. The input is the generated voice data, and the output is data transfer to the terminal. This is also done over the network.

[1427] Step 8:

[1428] The terminal plays the received voice data and lets the user and the other party listen. The terminal's input is the received voice data, and the output is the voice "Hello, what is your name?" played through the speaker.

[1429] Step 9:

[1430] The other person speaks in English into the device's microphone, "My name is John." This voice data is generated. The other person's input is an analog voice signal, and the device's output is digital voice data.

[1431] Step 10:

[1432] The terminal transmits the other party's digital voice data to the server. The input is the other party's digital voice data, and the output is the data transmitted to the server.

[1433] Step 11:

[1434] The server converts the received voice data back into text data using the Google Cloud Speech-to-Text API. The input is the other person's digital voice data, and the output is the text data "My name is John."

[1435] Step 12:

[1436] The server uses the Google Cloud Translation API to translate the text data into the specified language. The input is English text data, and the output is Japanese text data: "My name is John."

[1437] Step 13:

[1438] The server uses the Google Cloud Text-to-Speech API to convert the translated text data into audio data. The input is the Japanese text data "My name is John." and the output is audio data.

[1439] Step 14:

[1440] The server sends the generated voice data to the terminal. The input is the generated voice data, and the output is the data transfer to the terminal.

[1441] Step 15:

[1442] The terminal plays the received voice data and lets the user hear it. The input is the received voice data, and the output is the voice "My name is John" played through the speaker.

[1443] This allows for seamless real-time communication between users and speakers of other languages.

[1444] (Application example 1)

[1445] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1446] In conventional food delivery services, when the delivery person and the customer speak different languages, communication problems arise, making it difficult to deliver smoothly. This makes it difficult to immediately respond to the exact delivery location or special requests of the customer, and there is a need to improve the quality of the service.

[1447] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1448] In this invention, the server includes means for collecting voice data input by a user, means for transmitting the voice data to the server, means for the server to convert the voice data into character data, means for the server to translate the character data into a specified language, means for the server to convert the translated character data into voice data, means for outputting the translated voice data to the user, means for synthesizing the translation of the voice data into a specified language, and means for supporting communication between food delivery personnel and customers, thereby enabling smooth real-time communication between delivery personnel and customers who speak different languages.

[1449] "User-input voice data" refers to data that is collected by a device as an electrical signal when a user speaks through an input device such as a microphone.

[1450] "Means for transmitting to a server" refers to the general technology and protocols for transferring collected voice data to a server via a communications network.

[1451] "Means for converting into text data" refers to a process or engine that uses speech recognition technology to convert collected voice data into corresponding text data.

[1452] "Means for translating into a specified language" refers to technology for converting text data into a different language using a natural language processing engine.

[1453] "Means for converting into voice data" refers to a process for converting the translated text data into voice data using voice synthesis technology that synthesizes natural human speech from text.

[1454] "Means for outputting translated audio data to the user" refers to technology for playing the audio data through an output device such as a speaker.

[1455] "Means for supporting communication between food delivery personnel and customers" refers to all technologies that utilize a multilingual automatic translation system to enable real-time communication between delivery personnel and customers who speak different languages.

[1456] This invention is a multilingual automatic translation system that supports real-time communication between a user and people who speak different languages, and has an embodiment specialized for food delivery services. Specific processing and each means of this embodiment will be described below.

[1457] System Overview:

[1458] The system provides real-time multilingual translation between delivery personnel and customers, facilitating smooth communication. The system is realized by combining speech recognition, natural language processing, and speech synthesis technologies. The main hardware and software used include smartphones, servers, speech recognition engines (e.g., Google Speech-to-Text API), natural language processing engines (e.g., Google Translate API), and speech synthesis engines (e.g., gTTS).

[1459] Program processing:

[1460] Voice collection and recognition:

[1461] The device (smartphone) collects Japanese voice data input by the user. The device uses a built-in microphone to record the user's voice and sends it to the server as digital voice data. The server then uses the Google Speech-to-Text API to convert this voice data into text data.

[1462] Text translation:

[1463] The server translates the converted text data into a specified language, for example, English, using the Google Translate API. The translated text data is temporarily stored on the server.

[1464] Text-to-Speech:

[1465] The translated text data is converted into audio data using gTTS (Google Text-to-Speech), which is then sent back to the device and played through the speaker.

[1466] Examples:

[1467] User voice input:

[1468] For example, a delivery person might say in Japanese, "May I confirm your address?" The device collects this voice and sends it to the server.

[1469] Server process:

[1470] The speech recognition engine converts the speech into text and saves it in the log as "May I confirm your address?"

[1471] A natural language processing engine translates this text into English as "Can I confirm the address?"

[1472] A speech synthesis engine converts the translated text into speech.

[1473] Audio Output:

[1474] The device plays a translated English audio message that the delivery person can instantly understand in the customer's language, facilitating the confirmation of the appropriate delivery location.

[1475] Examples of prompts:

[1476] A user launches the app, sets "Japanese" as the primary language and "English" as the target language, and then says "May I confirm your address?" in Japanese. The app recognizes this, translates it into English, and plays it back aloud.

[1477] In this way, the barrier of multilingual communication in food delivery services is resolved, enabling smooth communication between delivery personnel and customers.

[1478] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1479] Step 1:

[1480] The user launches the app and provides voice input. The user launches the app on their smartphone and sets "Japanese" as the main language and "English" as the translation language. They then speak in Japanese, for example, "May I confirm your address?" The device's microphone collects this voice and processes it as digital voice data. The input in this step is the user's voice, and the output is digital voice data.

[1481] Step 2:

[1482] The device sends the collected audio data to the server. The device uploads this digital audio data to the server via an internet connection. The input in this step is the digital audio data, and the output is the audio data sent to the server.

[1483] Step 3:

[1484] The server uses a speech recognition engine to convert the transmitted voice data into text data. Specifically, the server calls the Google Speech-to-Text API to convert the voice data into text format. The input in this step is voice data, and the output is text data such as "May I confirm your address?"

[1485] Step 4:

[1486] The server passes the converted text data to a natural language processing engine and translates it into the specified language. Specifically, the server uses the Google Translate API to convert the text data "May I confirm the address?" into the English text "Can I confirm the address?". The input in this step is the text data, and the output is the translated text data.

[1487] Step 5:

[1488] The server uses a speech synthesis engine to convert the translated text data into audio data. Specifically, the server uses gTTS (Google Text-to-Speech) to convert the translated English text into an audio file. The input in this step is the translated text data, and the output is audio data.

[1489] Step 6:

[1490] The server sends the generated voice data to the terminal. The server uploads the generated voice data to the terminal via an internet connection. The input in this step is the voice data, and the output is the voice data sent to the terminal.

[1491] Step 7:

[1492] The device plays the received audio data. Specifically, it uses the device's speaker to play the translated English audio. By asking the user, "Can I confirm the address?", the delivery person can immediately understand it in the customer's language. The input in this step is audio data, and the output is audio that the user can hear.

[1493] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1494] The multilingual machine translation system of the present invention combines speech recognition, natural language processing, speech synthesis, and emotion recognition technologies to support real-time communication between users and speakers of other languages. Users input what they want to say using their smartphone, and the system instantly translates and outputs voice based on the emotion.

[1495] System Overview:

[1496] This system collects the user's voice data, performs speech recognition and emotion recognition on the server, and translates it using natural language processing. Furthermore, it synthesizes speech based on the emotion recognition results and outputs speech that reflects the appropriate emotion to the user. This not only overcomes language barriers, but also makes it possible to accurately convey emotions.

[1497] Specific steps for implementation:

[1498] User Action:

[1499] The user launches the app and sets the "language they speak," "language to convert to," "desired voice," and "emotion recognition settings" on the settings screen. The user then speaks into the smartphone's microphone, saying, for example, "Hello, what's your name?"

[1500] Audio data collection:

[1501] The terminal collects the user's voice data as a digital signal, converts it into voice data, and transmits it to a server via the Internet.

[1502] Speech and emotion recognition:

[1503] The server passes the received voice data to a voice recognition engine and an emotion engine. The voice recognition engine converts the voice data into text data, and the emotion engine recognizes the user's emotion (e.g., joy, sadness, anger, etc.).

[1504] Translation process:

[1505] The recognized text data is passed to a natural language processing engine and translated into the specified language. For example, "Hello, what is your name?" is translated into "Hello, what is your name?"

[1506] Emotion-based speech synthesis:

[1507] The server sends the translated text data and the recognized emotion data to a speech synthesis engine. The speech synthesis engine adjusts the translation result based on the emotion and generates speech data that reflects the appropriate emotion. For example, if the user is happy, the tone of the voice will also be adjusted to express happiness.

[1508] The resulting output:

[1509] The device receives the voice data from the server and plays it back through the smartphone speaker. The voice that is played back is "Hello, what is your name?", and reflects the user's emotions.

[1510] Response from the other party:

[1511] The other person responds in English, for example, "My name is John."

[1512] Processing the other party's voice data:

[1513] The device collects the other party's voice data and sends it back to the server. The server then performs speech recognition, natural language processing, and speech synthesis in the same way, and outputs the translated and emotion-reflected voice data to the user. For example, a voice saying "My name is John" is played back with the appropriate emotion.

[1514] Examples:

[1515] 1. User and partner initial settings

[1516] User: Launch the app and set "Japanese" as the speaking language, "English" as the target language, and "Male Voice" as the desired voice. Set emotion recognition to "On."

[1517] Other person: Speak in English.

[1518] 2. User utterances

[1519] User: Say "Hello, what's your name?"

[1520] Terminal: Sends audio data to the server.

[1521] Server: The speech recognition engine converts the speech into text data. It saves the message in the log as "Hello, what is your name?" The emotion engine recognizes "joy." The natural language processing engine translates it into "Hello, what is your name?" The speech synthesis engine converts it into English speech that expresses "joy." The speech data is sent to the device.

[1522] Device: Play the received audio. "Hello, what is your name?"

[1523] 3. The other person's response

[1524] Other person: Say, "My name is John."

[1525] Terminal: Sends the other party's voice data to the server.

[1526] Server: The speech recognition engine converts the speech into text data. It saves the text in the log as "My name is John." The emotion engine recognizes "neutral." The natural language processing engine translates it to "My name is John." The speech synthesis engine converts it into Japanese speech that includes "neutral." The speech data is sent to the device.

[1527] Device: Play the received audio: "My name is John."

[1528] This allows users and their partners to communicate seamlessly in real time, reflecting their emotions. This invention is a groundbreaking system in that it goes beyond simple language translation and also enables the transmission of emotions.

[1529] The processing flow will be explained below.

[1530] Step 1:

[1531] User: Launch the app and set the "language you speak," "language you want to convert to," "desired voice," and "emotion recognition settings" on the settings screen. For example, set "Japanese" as the language you speak, "English" as the language you want to convert to, "male voice" as the desired voice, and turn on "emotion recognition."

[1532] Step 2:

[1533] Terminal: Saves the user's settings and enters a standby state for voice input.

[1534] Step 3:

[1535] User: Begins speaking into the smartphone microphone. For example, "Hello, what's your name?"

[1536] Step 4:

[1537] Terminal: Collects the user's voice as a digital signal, converts it into voice data, and sends the converted voice data to a server via the Internet.

[1538] Step 5:

[1539] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "Hello, what's your name?". The converted text data is recorded in a log.

[1540] Step 6:

[1541] Server: Passes the text data to the natural language processing engine and translates it into the specified language. The natural language processing engine translates "Hello, what is your name?" into "Hello, what is your name?". The translation result is recorded in a log.

[1542] Step 7:

[1543] Server: Passes the voice data to the emotion engine and analyzes the user's emotion. The emotion engine generates an emotion tag from the voice, recognizing "joy" in this example, and records the result in a log.

[1544] Step 8:

[1545] Server: Passes the translated text data and emotion tag to the speech synthesis engine. The speech synthesis engine converts "Hello, what is your name?" into voice data in a "male voice" that reflects the emotion of "joy." The generated voice data is then prepared for transmission to the device.

[1546] Step 9:

[1547] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "Hello, what is your name?", with a tone that expresses "joy."

[1548] Step 10:

[1549] Other person: Respond in English. For example, say, "My name is John."

[1550] Step 11:

[1551] Terminal: Collects the other party's voice data again as a digital signal and sends it to a server via the Internet.

[1552] Step 12:

[1553] Server: Passes the received voice data to the speech recognition engine. The speech recognition engine converts the voice data into text data. The resulting text data is "My name is John." The converted text data is recorded in a log.

[1554] Step 13:

[1555] Server: Passes the text data to the natural language processing engine and translates it into the specified language. The natural language processing engine translates "My name is John." to "My name is John." The translation result is recorded in a log.

[1556] Step 14:

[1557] Server: Passes the voice data to the emotion engine and analyzes the other person's emotion. The emotion engine generates an emotion tag from the voice, recognizing "neutral" in this example. Records the result in a log.

[1558] Step 15:

[1559] Server: Passes the translated text data and emotion tag to the speech synthesis engine. The speech synthesis engine converts "My name is John" into voice data in a "male voice" that reflects a "neutral" emotion. The generated voice data is then prepared for transmission to the device.

[1560] Step 16:

[1561] Device: The voice data received from the server is imported and played from the smartphone speaker. The voice that is played is "My name is John," in a tone that expresses "neutrality."

[1562] This step enables users and other parties to overcome language barriers and achieve real-time communication that accurately conveys emotions.

[1563] Example 2

[1564] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1565] Conventional multilingual translation systems simply translate languages ​​and have difficulty conveying emotional nuances. As a result, they are unable to accurately convey emotions, which play an important role in communication, and are therefore less convenient for actual dialogue. Furthermore, they have the problem of being difficult to translate and reflect emotions in real time, which hinders smooth communication.

[1566] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1567] In this invention, the server includes a means for converting voice data into text data, a means for recognizing emotions in the text data, and a means for converting the translated text data into voice data that reflects the emotions, thereby enabling accurate conveyance of emotions and smooth real-time multilingual communication.

[1568] "Audio data" refers to sound information input by a user through the microphone of a terminal, converted into a digital signal.

[1569] A "server" is a central processing unit that receives voice data and performs various processes such as preprocessing, voice recognition, emotion recognition, translation, and voice synthesis.

[1570] "Character data" is data in text format that has been converted from voice data by a voice recognition engine.

[1571] "Translation" is the process of converting character data into a specified language.

[1572] "Emotion recognition" is a technology that analyzes and identifies emotions (joy, surprise, sadness, etc.) from a user's voice data.

[1573] "Speech synthesis" is a technology that artificially generates natural-sounding speech based on character data and emotion recognition results.

[1574] A "terminal" is a digital device used by a user, such as a smartphone or tablet.

[1575] The multilingual machine translation system of the present invention combines speech recognition, natural language processing, speech synthesis, and emotion recognition technologies to support real-time communication between users and speakers of other languages. This system collects speech data provided by users, translates it through processing on the server, and generates and outputs speech that reflects appropriate emotions.

[1576] Hardware and Software Configuration

[1577] The main components of the system are:

[1578] 1. Device (smartphone or tablet):

[1579] A microphone that collects audio data

[1580] A function that converts audio data into a digital signal

[1581] Ability to send data to a server via the Internet

[1582] A speaker that plays back audio data received from the server

[1583] 2. Server:

[1584] Speech recognition engine (e.g., Google Cloud Speech-to-Text): converts voice data into text data

[1585] Emotion recognition engine (e.g., IBM Watson Tone Analyzer): Analyzes emotions from voice data

[1586] Natural language processing engine (e.g., Google Cloud Translation API): Translates text data into a specified language

[1587] Speech synthesis engine (e.g., Amazon Polly): Generates speech based on translated text data and emotion data.

[1588] Examples of the invention

[1589] Example 1:

[1590] The user launches the app and configures the following settings on the settings screen:

[1591] "Languages ​​I speak": "Japanese"

[1592] "Language to convert to": "English"

[1593] "Voice of Hope": "Male Voice"

[1594] "Emotion Recognition Settings": "On"

[1595] Next, the user speaks into the smartphone's microphone, saying, "Look at that dog!" The device collects this voice data as a digital signal and sends it to a server via the Internet. On the server, a voice recognition engine converts the voice data into "text data" and generates the text data "Look at that dog!" The emotion recognition engine then recognizes the user's emotion as "surprise" from the text data.

[1596] The server's natural language processing engine translates the text data into the specified language (English) and obtains the translation result "Look at that dog!". The speech synthesis engine generates English speech that reflects the appropriate emotion based on this translation result and the emotion of "surprise." The server then sends the final generated speech data to the device, and the surprised voice "Look at that dog!" is played from the device's speaker.

[1597] Example 2:

[1598] When the other person responds with "Yeah, it's really cute!", the device converts the other person's voice data into a digital signal and sends it to the server. The server uses a speech recognition engine to convert "Yeah, it's really cute!" into text data, and an emotion recognition engine recognizes the emotion "neutral." The natural language processing engine translates this to "Yeah, it's really cute!", and a speech synthesis engine generates Japanese speech that reflects a neutral tone. Finally, the generated voice data is sent to the device, and the speaker plays back the voice "Yeah, it's really cute!"

[1599] Example prompt sentence:

[1600] "The user launches the app and sets Japanese as the language they speak and English as the translation language. The user says, "Hello, what is your name?" The voice data is sent to the server, the speech recognition engine converts it to text, and the emotion engine recognizes "joy." The text data is translated by the NLP engine into "Hello, what is your name?" and the speech synthesis engine generates English speech that reflects joy. The audio is then played back on the user's device."

[1601] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1602] Step 1:

[1603] The user launches the app and enters the language they speak, the language they want to translate into, the desired voice, and emotion recognition settings on the settings screen. Specifically, they set "Japanese" as the language they speak and "English" as the target language, then select a "male voice" and turn emotion recognition "on." Once the settings are complete, the user taps the "Save Settings" button.

[1604] Input: User settings

[1605] Output: The app with the settings saved

[1606] Step 2:

[1607] After completing the setup, the user speaks into the smartphone's microphone, for example, saying, "Look at that dog!" The device converts the voice data through the microphone into a digital signal and sends this data to a server via the Internet.

[1608] Input: User's voice

[1609] Output: Audio data sent to the server

[1610] Step 3:

[1611] The server receives the voice data and passes it to a speech recognition engine. The speech recognition engine (e.g., Google Cloud Speech-to-Text) converts the voice data into text data. For example, it generates text data such as "Look at that dog!" The server then passes the text data to an emotion recognition engine (e.g., IBM Watson Tone Analyzer), which recognizes the user's emotion as "surprise" from the text data.

[1612] Input: Audio data

[1613] Output: Text data and emotion data

[1614] Step 4:

[1615] The server passes the generated text data and emotion data to a natural language processing engine. The natural language processing engine (e.g., Google Cloud Translation API) translates this text data into the specified language (English in this case). For example, "Look at that dog!" is translated to "Look at that dog!"

[1616] Input: Text data and emotion data

[1617] Output: Translated text data (English)

[1618] Step 5:

[1619] The server sends the translated text data and emotion data to a speech synthesis engine. The speech synthesis engine (e.g., Amazon Polly) generates voice data that reflects the user's emotion based on the translation results. For example, voice data that expresses surprise and says, "Look at that dog!"

[1620] Input: Translated text data and emotion data

[1621] Output: Emotion-reflecting audio data

[1622] Step 6:

[1623] The server sends the generated voice data to the device, which then plays the received voice data through the smartphone speaker, allowing the user to hear the surprised voice saying, "Look at that dog!"

[1624] Input: Audio data

[1625] Output: Audio played through the speaker

[1626] Step 7:

[1627] The other person responds to the user's speech, for example, "Yeah, it's really cute!" The device collects the other person's voice data and sends it back to the server.

[1628] Input: Other party's voice

[1629] Output: Audio data sent to the server

[1630] Step 8:

[1631] The server then passes the received voice data back to the voice recognition engine, which converts it into text data. For example, the text data generated is "Yeah, it's really cute!" The server then passes the text data to the emotion recognition engine, which analyzes the emotion and determines that it is "neutral." The translation engine translates the text data into Japanese, obtaining the translation result "Yeah, it's really cute!" Finally, the speech synthesis engine generates Japanese speech based on this translation result and the "neutral" emotion. The device then plays back the received voice data, allowing the user to hear the voice saying "Yeah, it's really cute!"

[1632] Input: Voice data of the other party

[1633] Output: Emotionally-reflected Japanese speech data

[1634] (Application example 2)

[1635] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1636] In modern factories, it is becoming increasingly common for multinational workers to work together, which can lead to communication issues between workers who speak different languages. This can lead to reduced work efficiency and potential misunderstandings and work errors. Furthermore, in addition to simple language translation, there are also issues with emotions not being conveyed when giving work instructions or reporting, which can lead to a lack of understanding of the urgency and importance of instructions. A system that can solve these problems is needed.

[1637] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1638] In this invention, the server includes means for collecting voice data input by a user, means for transmitting the voice data to the server, means for converting the voice data into character data by the server, means for translating the character data into a specified language by the server, means for converting the translated character data into voice data by the server, means for recognizing emotions from the voice data and the character data by the server, means for adjusting the translated voice data based on emotions, and means for outputting the translated voice data to a user. This enables real-time communication between workers who speak different languages ​​that takes emotions into consideration.

[1639] "Voice data" refers to digital signals that collect end-user speech in audio form.

[1640] A "server" is a central computer system that processes, converts, and translates voice data over a network.

[1641] "Character data" is information in text format converted from voice data.

[1642] "Translation" is the process of converting text data from one language into another.

[1643] "Emotion recognition" is a technology that identifies a speaker's emotions (e.g., joy, sadness, anger, etc.) from audio data and text data.

[1644] "Speech synthesis" is a technology that generates speech based on text data.

[1645] "Real-time" means that processing occurs almost immediately, with minimal time delay.

[1646] "Modification" is the process of changing the tone and intonation of speech data based on perceived emotion.

[1647] "Log" means data storage for recording and saving processed data.

[1648] The present invention is a multilingual automatic translation system that translates what a user says into other languages ​​in real time and outputs the translated content while reflecting the user's emotions. Specific embodiments for carrying out the present invention will be described below.

[1649] Program Overview

[1650] The server is composed of multiple components, including a speech recognition engine, emotion recognition engine, translation engine, and speech synthesis engine. The user's device (for example, a smartphone or robot) collects voice data and sends it to the server. When the server receives the voice data, it performs speech recognition and converts it into text data. It then performs emotion recognition to identify the user's emotion. The translation engine translates the text data into the specified language, and the speech synthesis engine generates speech based on the translated text data and emotion data. Finally, the generated voice data is sent to the user's device and played back from the device.

[1651] Hardware and software used

[1652] 1. Hardware:

[1653] User terminal: devices such as smartphones and robots

[1654] Server: High-performance computing server

[1655] 2. Software:

[1656] Speech recognition engine: Google Cloud Speech-To-Text API

[1657] Emotion recognition engine: AWS Comprehend

[1658] Translation engine: Google Cloud Translation API

[1659] Speech synthesis engine: Google Cloud Text-To-Speech

[1660] Process flow and concrete examples

[1661] 1. User Action:

[1662] Users talk to their smartphones or robots, for example, a factory worker might say, "Put the new part here."

[1663] 2. Collection and Transmission of Audio Data:

[1664] The user device collects the voice data and sends it to a server via the Internet, where it is transmitted stably in digital format.

[1665] 3. Speech and Emotion Recognition:

[1666] The server receives the voice data and converts it into text data using a voice recognition engine. For example, voice data in Japanese saying "Please place the new part here" is converted into text data saying "Please place the new part here." Next, an emotion recognition engine analyzes the user's emotion from the text data and voice data and recognizes it as "neutral."

[1667] 4. Translation process:

[1668] The translation engine translates the text data into English and obtains the translation result "Please place the new part here."

[1669] 5. Speech synthesis:

[1670] The speech synthesis engine uses the translated text data and emotion data to generate a neutral voice saying, "Please place the new part here."

[1671] 6. Result output:

[1672] The generated voice data is sent to the user's device and played back through the device's speaker, enabling real-time, emotional communication between factory robots and workers who speak different languages.

[1673] Prompt Sentence Examples

[1674] A concrete example of a worker giving instructions to a robot:

[1675] Worker: Instructions in Japanese: "Put the new part here."

[1676] Robot: Outputs the instruction in English: "Please place the new part here."

[1677] In this way, the multilingual automatic translation system of the present invention realizes seamless communication that takes emotions into consideration between users who speak different languages.

[1678] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1679] Step 1:

[1680] A means of collecting user-input voice data

[1681] The user speaks into the microphone of their smartphone or robot, and the voice data is collected and converted into a digital signal, which serves as input for subsequent processing.

[1682] Step 2:

[1683] A means of sending audio data to the server

[1684] The device sends the collected audio data to a server via the internet. The audio data is sent to the server using a secure communication protocol (e.g. HTTPS). The output at this stage is digital audio data sent to the server.

[1685] Step 3:

[1686] A means of performing voice recognition and converting voice data into text data

[1687] The server passes the received voice data to a speech recognition engine (for example, Google Cloud Speech-To-Text API). The speech recognition engine analyzes the voice data and converts it into text data. The output is text data in text format, such as "Hello."

[1688] Step 4:

[1689] A means of performing emotion recognition and identifying the user's emotions

[1690] The server passes the text data and the original audio data to an emotion recognition engine (e.g., AWS Comprehend). The emotion recognition engine analyzes the text and tone of the audio to identify the user's emotion (e.g., happy, neutral, angry, etc.). The output is an emotion tag (e.g., "happy").

[1691] Step 5:

[1692] A means of translating character data into a specified language

[1693] The server passes the text data to a translation engine (e.g., Google Cloud Translation API). The translation engine translates the text data into the specified language. The output is the translated text data (e.g., "Hello").

[1694] Step 6:

[1695] A means of converting translated text data into emotionally adjusted speech data

[1696] The server passes the translated text data and emotion data to a speech synthesis engine (such as Google Cloud Text-To-Speech). The speech synthesis engine converts the text data into speech data, generating speech with intonation and tone that reflects the recognized emotion. The output is speech data adjusted based on the emotion.

[1697] Step 7:

[1698] A means for outputting audio data to a user's terminal

[1699] The server then sends the generated voice data to the user's device, which then plays the received voice data and outputs the voice from its speaker. This allows the user's interlocutor (e.g., a factory robot) to receive instructions and information in a voice that reflects the appropriate emotion.

[1700] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1701] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1702] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1703] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1704] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1705] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1706] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1707] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1708] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1709] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1710] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1711] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1712] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1713] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1714] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1715] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1716] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1717] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1718] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1719] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1720] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1721] The following is further disclosed regarding the above embodiment.

[1722] (Claim 1)

[1723] means for collecting voice data input by a user;

[1724] means for transmitting the voice data to a server;

[1725] means for converting the voice data into character data by the server;

[1726] means for translating the character data into a language designated by the server;

[1727] means for converting the translated text data into audio data by the server;

[1728] means for outputting the translated speech data to a user;

[1729] A system including:

[1730] (Claim 2)

[1731] 2. The system according to claim 1, further comprising means for collecting and transmitting voice data input by the user in real time.

[1732] (Claim 3)

[1733] 2. The system of claim 1, wherein the server comprises means for storing the audio data in a log in a translatable format.

[1734] "Example 1"

[1735] (Claim 1)

[1736] means for collecting voice data input by a user;

[1737] means for transmitting the voice data to a server;

[1738] means for converting the voice data into character data by the server;

[1739] means for translating the character data into a language designated by the server;

[1740] means for converting the translated text data into audio data by the server;

[1741] means for outputting the translated speech data to a user;

[1742] means for converting the voice data into digital data before transmitting the voice data;

[1743] means for generating voice data in a voice desired by the user;

[1744] means for transmitting the generated voice data to a terminal;

[1745] A system including:

[1746] (Claim 2)

[1747] 2. The system according to claim 1, further comprising means for collecting and transmitting voice data input by the user in real time.

[1748] (Claim 3)

[1749] 2. The system of claim 1, wherein the server comprises means for storing the audio data in a log in a translatable format.

[1750] "Application Example 1"

[1751] (Claim 1)

[1752] means for collecting voice data input by a user;

[1753] means for transmitting the voice data to a server;

[1754] means for converting the voice data into character data by the server;

[1755] means for translating the character data into a language designated by the server;

[1756] means for converting the translated text data into audio data by the server;

[1757] means for outputting the translated speech data to a user;

[1758] means for synthesizing the translation of the voice data into a designated language;

[1759] A means of supporting communication between food delivery staff and customers,

[1760] A system including:

[1761] (Claim 2)

[1762] 2. The system according to claim 1, further comprising means for collecting and transmitting voice data input by the user in real time, and translating the data so that the delivery person can immediately understand the data in the customer's language.

[1763] (Claim 3)

[1764] 2. The system according to claim 1, further comprising means for storing the voice data in a log in a translatable format and for maintaining a communication history between the delivery person and the customer.

[1765] "Example 2: Combining Emotion Engines"

[1766] (Claim 1)

[1767] means for collecting voice data input by a user;

[1768] means for transmitting the voice data to a server;

[1769] means for converting the voice data into character data by the server;

[1770] means for translating the character data into a language designated by the server;

[1771] means for the server to recognize emotions in the character data;

[1772] means for converting the translated text data into voice data that reflects emotions by the server;

[1773] means for outputting the translated speech data to a user;

[1774] A system including:

[1775] (Claim 2)

[1776] 2. The system according to claim 1, further comprising means for collecting and transmitting voice data input by the user in real time.

[1777] (Claim 3)

[1778] 2. The system of claim 1, wherein the server comprises means for storing the audio data in a log in a translatable format.

[1779] "Application example 2 when combining emotion engines"

[1780] (Claim 1)

[1781] means for collecting voice data input by a user;

[1782] means for transmitting the voice data to a server;

[1783] means for converting the voice data into character data by the server;

[1784] means for translating the character data into a language designated by the server;

[1785] means for converting the translated text data into audio data by the server;

[1786] means for recognizing emotions from the voice data and the character data in the server;

[1787] means for adjusting the translated speech data based on emotion;

[1788] means for outputting the translated speech data to a user;

[1789] A system including:

[1790] (Claim 2)

[1791] 2. The system according to claim 1, further comprising means for collecting, emotion-recognizing, and transmitting voice data input by the user in real time.

[1792] (Claim 3)

[1793] 2. The system of claim 1, wherein the server stores the speech data in a log in a translatable format and further comprises means for recording recognized emotion information. [Explanation of symbols]

[1794] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for collecting voice data input by a user; means for transmitting the voice data to a server; means for converting the voice data into character data by the server; means for translating the character data into a language designated by the server; means for converting the translated text data into audio data by the server; means for outputting the translated speech data to a user; A system including:

2. 2. The system according to claim 1, further comprising means for collecting and transmitting voice data input by the user in real time.

3. 2. The system of claim 1, wherein the server comprises means for storing the audio data in a log in a translatable format.

Images